加入实验输出(批次 13/14)

This commit is contained in:
2026-09-23 23:37:34 +08:00
parent 0c9c62de03
commit 6f0c9cfce7
91 changed files with 44382 additions and 0 deletions
@@ -0,0 +1,5 @@
# TSFA five-fold result bundle
Seven frozen-alignment comparisons on 100 held-out clips from 37 video IDs. All Python runs use the project uv environment. Read `run_manifest.json` for the complete protocol and limitations. `paired_contrasts.csv` reports left minus right with video-ID-cluster bootstrap confidence intervals. The full `outputs/tsfa/` directory retains per-clip CSVs and probe checkpoints.
The `alignment_only_*` summaries pool identical train-fold-standardized raw source features through each method's alignment matrix. They isolate source selection from native model value/output projections. `tsfa_temporal_diagnostics_summary.csv` reports MVR, span, entropy, and source coverage.
@@ -0,0 +1,8 @@
method,clip_count,video_id_count,content_auc_mean_video_macro_mean,content_auc_mean_ci95_low,content_auc_mean_ci95_high,canonical_pairwise_time_mae_mean_video_macro_mean,canonical_pairwise_time_mae_mean_ci95_low,canonical_pairwise_time_mae_mean_ci95_high,temporal_quality_1_minus_mae_video_macro_mean,temporal_quality_1_minus_mae_ci95_low,temporal_quality_1_minus_mae_ci95_high
M3_noSourceTime,100,37,0.7079247666486141,0.6981638469348909,0.7180669733515811,0.25125551939830126,0.24533393078354204,0.25730599312459385,0.7487444806016986,0.7429385193325527,0.7543707643381519
M3_sourceTime,100,37,0.5646895233632823,0.5540910611345894,0.5751509210736825,0.09874174948688212,0.08898600889351087,0.10906350604411601,0.9012582505131179,0.8909351143570293,0.9111176736631803
M4_sourceTime,100,37,0.5443730644271425,0.5314674094814167,0.556163600543299,0.06439514300291252,0.05872951511472842,0.069915719671434,0.9356048569970875,0.929956941099165,0.9411986290102992
TSFA-main,100,37,0.6265827216844504,0.6161616236415937,0.6364439680609562,0.0703286687392714,0.06460036690688006,0.07572429366463695,0.9296713312607287,0.9244517958993883,0.9352479536843178
TSFA-multiply,100,37,0.6266223446090468,0.6163471959183728,0.6367020258052505,0.06779188350752184,0.06251919458961656,0.07342671925523597,0.932208116492478,0.9266964480620523,0.9376652354766101
TSFA-random,100,37,0.6317354999281196,0.6245134023121242,0.6386649427182931,0.25069943769545594,0.2462397150003418,0.2551459569730119,0.7493005623045442,0.7451492641632237,0.7535130139973931
TSFA-global,100,37,0.6850667176168063,0.6715353281663122,0.6991479845315172,0.2432284738286704,0.23726117654439974,0.24953340198194693,0.7567715261713297,0.7508184770882113,0.7626622731375607
1 method clip_count video_id_count content_auc_mean_video_macro_mean content_auc_mean_ci95_low content_auc_mean_ci95_high canonical_pairwise_time_mae_mean_video_macro_mean canonical_pairwise_time_mae_mean_ci95_low canonical_pairwise_time_mae_mean_ci95_high temporal_quality_1_minus_mae_video_macro_mean temporal_quality_1_minus_mae_ci95_low temporal_quality_1_minus_mae_ci95_high
2 M3_noSourceTime 100 37 0.7079247666486141 0.6981638469348909 0.7180669733515811 0.25125551939830126 0.24533393078354204 0.25730599312459385 0.7487444806016986 0.7429385193325527 0.7543707643381519
3 M3_sourceTime 100 37 0.5646895233632823 0.5540910611345894 0.5751509210736825 0.09874174948688212 0.08898600889351087 0.10906350604411601 0.9012582505131179 0.8909351143570293 0.9111176736631803
4 M4_sourceTime 100 37 0.5443730644271425 0.5314674094814167 0.556163600543299 0.06439514300291252 0.05872951511472842 0.069915719671434 0.9356048569970875 0.929956941099165 0.9411986290102992
5 TSFA-main 100 37 0.6265827216844504 0.6161616236415937 0.6364439680609562 0.0703286687392714 0.06460036690688006 0.07572429366463695 0.9296713312607287 0.9244517958993883 0.9352479536843178
6 TSFA-multiply 100 37 0.6266223446090468 0.6163471959183728 0.6367020258052505 0.06779188350752184 0.06251919458961656 0.07342671925523597 0.932208116492478 0.9266964480620523 0.9376652354766101
7 TSFA-random 100 37 0.6317354999281196 0.6245134023121242 0.6386649427182931 0.25069943769545594 0.2462397150003418 0.2551459569730119 0.7493005623045442 0.7451492641632237 0.7535130139973931
8 TSFA-global 100 37 0.6850667176168063 0.6715353281663122 0.6991479845315172 0.2432284738286704 0.23726117654439974 0.24953340198194693 0.7567715261713297 0.7508184770882113 0.7626622731375607
@@ -0,0 +1,8 @@
method,clip_count,video_id_count,content_auc_mean_video_macro_mean,content_auc_mean_ci95_low,content_auc_mean_ci95_high,canonical_pairwise_time_mae_mean_video_macro_mean,canonical_pairwise_time_mae_mean_ci95_low,canonical_pairwise_time_mae_mean_ci95_high,temporal_quality_1_minus_mae_video_macro_mean,temporal_quality_1_minus_mae_ci95_low,temporal_quality_1_minus_mae_ci95_high
M3_noSourceTime,100,37,0.5915240976792394,0.5846003467230528,0.5986003527463633,0.25125551939830126,0.24526752948109784,0.2569629667175068,0.7487444806016986,0.7426625677892807,0.7546333254224017
M3_sourceTime,100,37,0.5159307443208153,0.5093207601202393,0.5227730292887872,0.09874174948688212,0.08938651093211228,0.10879260214913458,0.9012582505131179,0.890805973592594,0.9110025699716944
M4_sourceTime,100,37,0.5172970388067906,0.5080581473982847,0.5275570833339672,0.06439514300291252,0.0587335791039206,0.06983744635995676,0.9356048569970875,0.9296541211962468,0.9408869116126053
TSFA-main,100,37,0.5527356591964304,0.5398139072416833,0.5666946734419639,0.0703286687392714,0.06519967303243404,0.07598739113740521,0.9296713312607287,0.9244119213095845,0.9350019389370183
TSFA-multiply,100,37,0.5506013033677041,0.5379239892152792,0.5636824578678856,0.06779188350752184,0.06247147901094175,0.07313585563909683,0.932208116492478,0.926778081557954,0.9373936058949076
TSFA-random,100,37,0.5711638231839362,0.562858188378229,0.5801847256875272,0.25069943769545594,0.2462381631727499,0.25519553085514896,0.7493005623045442,0.7450096911264995,0.753607390700797
TSFA-global,100,37,0.6077635519257857,0.5944803976516411,0.6218732946519847,0.2432284738286704,0.2374573657121255,0.2489871789332721,0.7567715261713297,0.7506538401520033,0.7624564267860716
1 method clip_count video_id_count content_auc_mean_video_macro_mean content_auc_mean_ci95_low content_auc_mean_ci95_high canonical_pairwise_time_mae_mean_video_macro_mean canonical_pairwise_time_mae_mean_ci95_low canonical_pairwise_time_mae_mean_ci95_high temporal_quality_1_minus_mae_video_macro_mean temporal_quality_1_minus_mae_ci95_low temporal_quality_1_minus_mae_ci95_high
2 M3_noSourceTime 100 37 0.5915240976792394 0.5846003467230528 0.5986003527463633 0.25125551939830126 0.24526752948109784 0.2569629667175068 0.7487444806016986 0.7426625677892807 0.7546333254224017
3 M3_sourceTime 100 37 0.5159307443208153 0.5093207601202393 0.5227730292887872 0.09874174948688212 0.08938651093211228 0.10879260214913458 0.9012582505131179 0.890805973592594 0.9110025699716944
4 M4_sourceTime 100 37 0.5172970388067906 0.5080581473982847 0.5275570833339672 0.06439514300291252 0.0587335791039206 0.06983744635995676 0.9356048569970875 0.9296541211962468 0.9408869116126053
5 TSFA-main 100 37 0.5527356591964304 0.5398139072416833 0.5666946734419639 0.0703286687392714 0.06519967303243404 0.07598739113740521 0.9296713312607287 0.9244119213095845 0.9350019389370183
6 TSFA-multiply 100 37 0.5506013033677041 0.5379239892152792 0.5636824578678856 0.06779188350752184 0.06247147901094175 0.07313585563909683 0.932208116492478 0.926778081557954 0.9373936058949076
7 TSFA-random 100 37 0.5711638231839362 0.562858188378229 0.5801847256875272 0.25069943769545594 0.2462381631727499 0.25519553085514896 0.7493005623045442 0.7450096911264995 0.753607390700797
8 TSFA-global 100 37 0.6077635519257857 0.5944803976516411 0.6218732946519847 0.2432284738286704 0.2374573657121255 0.2489871789332721 0.7567715261713297 0.7506538401520033 0.7624564267860716
@@ -0,0 +1,22 @@
method,pair,clip_count,draw_count,same_time_similarity_video_macro_mean,same_time_similarity_ci95_low,same_time_similarity_ci95_high,video_id_count,shifted_far_similarity_video_macro_mean,shifted_far_similarity_ci95_low,shifted_far_similarity_ci95_high,same_minus_shifted_margin_video_macro_mean,same_minus_shifted_margin_ci95_low,same_minus_shifted_margin_ci95_high,matched_vs_shifted_auc_video_macro_mean,matched_vs_shifted_auc_ci95_low,matched_vs_shifted_auc_ci95_high,exact_r1_audio_to_vision_video_macro_mean,exact_r1_audio_to_vision_ci95_low,exact_r1_audio_to_vision_ci95_high,within_pm1_r1_audio_to_vision_video_macro_mean,within_pm1_r1_audio_to_vision_ci95_low,within_pm1_r1_audio_to_vision_ci95_high,mase_slots_audio_to_vision_video_macro_mean,mase_slots_audio_to_vision_ci95_low,mase_slots_audio_to_vision_ci95_high,exact_r1_vision_to_audio_video_macro_mean,exact_r1_vision_to_audio_ci95_low,exact_r1_vision_to_audio_ci95_high,within_pm1_r1_vision_to_audio_video_macro_mean,within_pm1_r1_vision_to_audio_ci95_low,within_pm1_r1_vision_to_audio_ci95_high,mase_slots_vision_to_audio_video_macro_mean,mase_slots_vision_to_audio_ci95_low,mase_slots_vision_to_audio_ci95_high,exact_r1_text_to_audio_video_macro_mean,exact_r1_text_to_audio_ci95_low,exact_r1_text_to_audio_ci95_high,within_pm1_r1_text_to_audio_video_macro_mean,within_pm1_r1_text_to_audio_ci95_low,within_pm1_r1_text_to_audio_ci95_high,mase_slots_text_to_audio_video_macro_mean,mase_slots_text_to_audio_ci95_low,mase_slots_text_to_audio_ci95_high,exact_r1_audio_to_text_video_macro_mean,exact_r1_audio_to_text_ci95_low,exact_r1_audio_to_text_ci95_high,within_pm1_r1_audio_to_text_video_macro_mean,within_pm1_r1_audio_to_text_ci95_low,within_pm1_r1_audio_to_text_ci95_high,mase_slots_audio_to_text_video_macro_mean,mase_slots_audio_to_text_ci95_low,mase_slots_audio_to_text_ci95_high,exact_r1_text_to_vision_video_macro_mean,exact_r1_text_to_vision_ci95_low,exact_r1_text_to_vision_ci95_high,within_pm1_r1_text_to_vision_video_macro_mean,within_pm1_r1_text_to_vision_ci95_low,within_pm1_r1_text_to_vision_ci95_high,mase_slots_text_to_vision_video_macro_mean,mase_slots_text_to_vision_ci95_low,mase_slots_text_to_vision_ci95_high,exact_r1_vision_to_text_video_macro_mean,exact_r1_vision_to_text_ci95_low,exact_r1_vision_to_text_ci95_high,within_pm1_r1_vision_to_text_video_macro_mean,within_pm1_r1_vision_to_text_ci95_low,within_pm1_r1_vision_to_text_ci95_high,mase_slots_vision_to_text_video_macro_mean,mase_slots_vision_to_text_ci95_low,mase_slots_vision_to_text_ci95_high
M3_noSourceTime,audio_vision,100,1.0,0.02983147300783216,-0.00780859958706209,0.06532931063779485,37,0.026275558042943677,-0.0135765906084752,0.06442367880532844,0.0035559149673156645,0.0011573716778813968,0.0058081254094310905,0.514271623637315,0.5086281381400399,0.5205522070322601,0.02223723723723724,0.019318307593307594,0.0252463963963964,0.0629030459030459,0.056346739596739595,0.0701120549120549,16.06521836121836,15.360586422136418,16.831551512226515,0.02093822393822394,0.01916058558558559,0.022803474903474903,0.0588974688974689,0.05515479407979409,0.06256855426855427,17.3258314028314,16.454282078507074,18.303939736164736,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_noSourceTime,text_audio,100,1.0,0.20791600453897077,0.18457893455935243,0.23068674583726764,37,0.0660212989394217,0.04607872245037598,0.08693206487027433,0.14189470507476618,0.1258566735337873,0.158192415668532,0.7246991238959324,0.7032607291086682,0.7463028065299342,,,,,,,,,,,,,,,,,,,0.13955383955383954,0.12312348777348776,0.156250858000858,0.27616302016302025,0.24729871299871298,0.30469959244959244,10.11119390819391,9.083786154011154,11.220066763191763,0.07386400686400688,0.06408464178464178,0.08439125911625915,0.16222651222651222,0.1441163770913771,0.18078545688545689,12.965773916773918,12.10177589017589,13.855149796224795,,,,,,,,,,,,,,,,,,
M3_noSourceTime,text_vision,100,1.0,0.032579638831090094,0.012203289357956712,0.053855902535509696,37,0.020862086972797865,-0.0005364448539057191,0.04502733759713527,0.01171755187282541,0.007967536814828653,0.015164253928682914,0.535601545504471,0.5297980337765578,0.5417672611678928,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.04206177606177607,0.03788120978120979,0.04638886743886744,0.10588416988416989,0.09782853925353927,0.113992277992278,14.162812097812097,13.519815894465893,14.831819369369372,0.023044616044616047,0.02111688116688117,0.025180180180180183,0.06238695838695839,0.05710733590733591,0.06806795366795368,17.50512955812956,16.450267953667954,18.578393522093524
M3_sourceTime,audio_vision,100,1.0,0.007696060061626377,-0.029554258849169186,0.04545826630162894,37,0.008636827414143384,-0.030810690011533682,0.04782087975048893,-0.0009407673630797903,-0.003493066123958016,0.0017008874928592713,0.5026440831028598,0.49047663000196245,0.5166576717342342,0.021732303732303733,0.018566891891891892,0.02489736164736165,0.05924495924495925,0.05102570785070786,0.06822308022308023,18.418811239811237,17.145313781638777,19.668025257400256,0.01908751608751609,0.017070527670527674,0.021123584298584297,0.05492621192621192,0.05029851994851995,0.05956598026598027,18.342253110253115,17.42752445302445,19.27038470613471,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_sourceTime,text_audio,100,1.0,0.030779919639574976,0.009495005332440814,0.051172133170129974,37,0.012129069987308123,-0.008134753400298974,0.03220063209724045,0.018650849718729958,0.01209877043793774,0.025570886226139,0.5453406619005025,0.5310778723489828,0.5597143654149971,,,,,,,,,,,,,,,,,,,0.03846546546546547,0.03189687902187902,0.04500438652938652,0.0974937794937795,0.08529543114543116,0.1109557915057915,16.05845045045045,15.243514607464608,16.823768436293438,0.026703560703560704,0.022845398970398972,0.030940111540111538,0.07566495066495067,0.06590900900900902,0.08547635135135136,15.480417417417419,14.34030947018447,16.63520718575719,,,,,,,,,,,,,,,,,,
M3_sourceTime,text_vision,100,1.0,0.0147220734507245,-0.008115254084484056,0.03707853268159794,37,0.014395209505575931,-0.008901017726582194,0.03943286969561065,0.0003268639452983987,-0.002732993650720293,0.0035072161571967925,0.4998074879590837,0.4950654246809501,0.504675911331164,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.020641784641784643,0.017091269841269845,0.024615486915486918,0.05474774774774775,0.04679173101673103,0.0628289897039897,19.39496525096525,18.298798712998718,20.349694851994855,0.020431574431574433,0.01848766623766624,0.02242575075075075,0.05971514371514372,0.05510990990990991,0.06392915057915058,16.647335478335478,15.535322372372372,17.931235296010293
M4_sourceTime,audio_vision,100,1.0,0.017589288202760855,-0.021384537387894743,0.05702271457169135,37,0.018760167504660787,-0.01933644305751479,0.05750124747004294,-0.0011708792981242977,-0.004425662602822854,0.0016943904201328112,0.5092526072805329,0.49760200300984875,0.5208348244378497,0.018691977691977692,0.014877273702273703,0.02274350064350065,0.04963020163020163,0.041014339339339345,0.05834750107250107,18.292485628485622,16.987942664092664,19.485725182325183,0.02013041613041613,0.017474710424710424,0.022690025740025738,0.05348219648219649,0.04821073573573574,0.05895694980694981,18.10186143286143,16.83444991419991,19.40507165379665,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M4_sourceTime,text_audio,100,1.0,0.02029622882045845,-0.015886606029959244,0.052645537742046604,37,0.004637936173545868,-0.02838808700016482,0.03639667112261427,0.01565829274056424,0.009190044274264193,0.022804289586498057,0.5389472080363038,0.514210176409312,0.5634661651963513,,,,,,,,,,,,,,,,,,,0.027861861861861865,0.02250713213213213,0.033694744744744744,0.06838953238953241,0.05814036894036893,0.07883789146289145,18.20780223080223,16.355241891891897,20.29683777348777,0.030084513084513088,0.02537934362934363,0.03572420634920635,0.07645002145002144,0.06447318747318749,0.09150033247533246,17.343640926640923,16.01122001287001,18.94879967824968,,,,,,,,,,,,,,,,,,
M4_sourceTime,text_vision,100,1.0,0.033176119103082845,-0.0041519938906857895,0.07124988025892566,37,0.030836823717937716,-0.008539991982258287,0.0710771330847214,0.0023392953867183023,-0.001506538955824979,0.006041902250845819,0.5036913011035352,0.4923280641013353,0.5162466670602972,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.018344916344916345,0.015335810810810813,0.021540604890604893,0.05032947232947234,0.042268693693693696,0.05863465250965252,18.6358845988846,17.37207688760189,20.017716023166024,0.020106392106392104,0.017946728871728875,0.022438331188331193,0.051333762333762335,0.045313084513084516,0.057213953238953244,19.317567996567995,17.859128196053195,20.795619787644785
TSFA-global,audio_vision,100,1.0,0.06059821270500538,0.023914981549968812,0.09491150040596706,37,0.03732938954030518,0.0038612152410352285,0.07181022922532529,0.02326882317339013,0.016046146106362763,0.031464541216613816,0.5711938543786947,0.5580918275513356,0.5851307426062079,0.05056928356928358,0.04562037752037753,0.055837269412269415,0.13590690690690693,0.12514061561561562,0.1478566066066066,13.252549978549979,12.300325911625912,14.232603314028314,0.02580952380952381,0.022693564993564997,0.02993452380952381,0.07257400257400258,0.06430289575289577,0.08383320463320464,16.92865637065637,15.890422136422137,17.992037709137712,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-global,text_audio,100,1.0,0.21948498332128252,0.18447166931635653,0.25264341574445714,37,0.13784915199225622,0.10805201589136158,0.16772503723338508,0.0816358312123013,0.0694722462373281,0.09469795330548046,0.6997278072145092,0.6772065253418312,0.7246179070170095,,,,,,,,,,,,,,,,,,,0.05226340626340627,0.043850450450450454,0.061542020592020574,0.1483732303732304,0.12583120978120982,0.1716539897039897,10.44393564993565,9.248472972972973,11.684178474903476,0.050666237666237665,0.04418186400686401,0.057607464607464604,0.13168897468897467,0.11565889103389104,0.1511234341484341,13.042119691119693,11.967449710424711,14.109006949806952,,,,,,,,,,,,,,,,,,
TSFA-global,text_vision,100,1.0,0.03882392940236514,-0.005101331978814988,0.08023915360506113,37,0.026369558634260028,-0.019257361349713748,0.0695494351361847,0.012454370779342111,0.006579723199721584,0.018851628858371938,0.5523689941841539,0.5389573468177523,0.5678277126290424,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.036459888459888457,0.032251780351780356,0.040649946374946386,0.10280694980694982,0.09098980051480052,0.11672229729729731,13.241670098670095,12.166959716859719,14.325898702273703,0.02583869583869584,0.02256068211068211,0.030118897468897468,0.06330373230373229,0.05587844272844273,0.07113738738738737,18.464465465465462,17.013464017589012,19.91242631917632
TSFA-main,audio_vision,100,1.0,0.013561541277063027,-0.01457795284016425,0.04262594169249218,37,0.008980510418424372,-0.021758340514473864,0.04004785757980806,0.004581030828024072,-0.0002248605515836633,0.010127997422592447,0.5114767559162506,0.5016843428677538,0.5212400272042559,0.02182110682110682,0.017747586872586874,0.02623365508365508,0.061913770913770906,0.05110411840411841,0.07329221364221364,17.73442985842986,16.826004118404118,18.77729813384814,0.022406692406692408,0.019272361647361645,0.026145913770913774,0.06514157014157014,0.056082850707850705,0.07448922136422137,15.752643071643071,14.584650504075505,16.82486615186615,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-main,text_audio,100,1.0,0.11990332463810639,0.0898675046922316,0.15305510010030474,37,0.07596570038282524,0.04989764804945704,0.10443112074815332,0.04393762438637949,0.03607569399696766,0.052024387539309866,0.6394096191556298,0.612673730897501,0.667128455450131,,,,,,,,,,,,,,,,,,,0.043396825396825406,0.03658292578292579,0.05185294937794939,0.12236121836121837,0.10395112612612614,0.14359046546546542,12.342245388245388,11.084742546117546,13.683040540540542,0.039102960102960105,0.03286315958815959,0.04523687258687259,0.10152981552981553,0.08593539253539254,0.11686853281853281,14.241285285285283,13.037042556842557,15.514015894465896,,,,,,,,,,,,,,,,,,
TSFA-main,text_vision,100,1.0,0.04098847119260702,0.005582397132704427,0.07929293901031172,37,0.037891365960320264,0.0005784316926466229,0.07758778326306248,0.0030971052297696644,-0.000814157517568749,0.0067062238179970675,0.5073206025174111,0.49017726806879075,0.5255564948211889,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.018722436722436724,0.014964897039897044,0.022390036465036463,0.05502488202488203,0.04573553196053196,0.06428104890604891,18.6562737022737,17.207105051480053,20.233926383526384,0.020561561561561564,0.01860630630630631,0.022591741741741746,0.05686486486486487,0.05081441441441443,0.06266833976833977,18.087792363792367,16.792936625911626,19.410561314886316
TSFA-multiply,audio_vision,100,1.0,0.00978179982461232,-0.02070967490279534,0.0421130031397444,37,0.0060832866483517,-0.02193733706144022,0.03656021604936627,0.003698513197226362,-0.0011122262813486495,0.009267909918290072,0.5129411883387948,0.5027842064281226,0.523200954159897,0.022283998283998283,0.017675825825825827,0.027219133419133415,0.05865551265551267,0.047279225654225654,0.07083616473616473,17.81548777348777,16.808302638352636,18.797270334620336,0.021042900042900044,0.01781427498927499,0.024212129987129988,0.0629056199056199,0.05411741741741742,0.07141601244101244,15.856422565422568,14.7419083976834,16.96319958172458,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-multiply,text_audio,100,1.0,0.11991994317142747,0.0889362489426478,0.15110231629245585,37,0.07792934440600745,0.052977172014714184,0.10278047049072589,0.04199059878088213,0.034127040037302034,0.05013125747396997,0.6336595358656528,0.6087905367706232,0.6589064419710364,,,,,,,,,,,,,,,,,,,0.04137323037323038,0.034712870012870024,0.048116763191763195,0.11405105105105105,0.09659961389961391,0.13316510081510083,12.645648219648223,11.431381585156586,13.97510311025311,0.04315615615615616,0.03623397683397684,0.05003088803088803,0.1083843843843844,0.09482691977691977,0.12210212355212356,14.16602059202059,13.052271911196911,15.273814092664091,,,,,,,,,,,,,,,,,,
TSFA-multiply,text_vision,100,1.0,0.03612421115527528,0.003679617927233912,0.0706390626773947,37,0.033054201762203535,0.0003160863252284142,0.06742515595895551,0.0030700093951693155,-0.0007710019667443696,0.006867931876338059,0.5052031858986646,0.4900221838004418,0.5219854911997864,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.01671685971685972,0.013541377091377095,0.019720752895752897,0.04806392106392106,0.03873852423852424,0.05764617117117116,19.210052767052765,17.75446628056628,20.751913610038613,0.021712569712569713,0.019381831831831835,0.02410126555126555,0.05798712998712999,0.05170907335907336,0.06397966537966537,18.30359202059202,16.94032631917632,19.627432818532814
TSFA-random,audio_vision,100,20.0,0.020861894386615707,0.005693970997846327,0.037061795072022896,37,0.01680333015403859,0.0004332554102972204,0.03303836082632705,0.0040585642224212435,0.0005754381947093255,0.007718638619050609,0.5193708956724581,0.5150352463917897,0.5238890623406223,0.03257623337623338,0.030417548262548257,0.0346354971042471,0.09581681681681682,0.08960499839124837,0.10207584298584299,15.154294980694983,14.546768505469755,15.738201547619045,0.021379450879450878,0.02055578024453025,0.022163257722007723,0.06266823251823253,0.060993607357357364,0.0645759614972115,16.373181810381816,15.610742067245818,17.204955935220937,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-random,text_audio,100,20.0,0.17050077302726946,0.13941037026448067,0.1990060504478285,37,0.09672763162535135,0.07223268336231446,0.12007034012783693,0.07377314134437793,0.06320199140044352,0.08493496876685644,0.6780049539566163,0.6551398515962296,0.7010971646109445,,,,,,,,,,,,,,,,,,,0.048387923637923644,0.04305602799227799,0.05432789575289575,0.13999991419991417,0.12458882453882453,0.15700594272844276,11.147006499356502,10.136636986808236,12.213333453989703,0.04792293007293008,0.04362501340626341,0.051922379343629345,0.12110746460746463,0.11033633526383527,0.13127376447876446,13.276559244959246,12.25213515122265,14.3009197506435,,,,,,,,,,,,,,,,,,
TSFA-random,text_vision,100,20.0,0.028480983388760747,-0.0010045256113203147,0.05799119457573753,37,0.026108679984648473,-0.004959284461738992,0.055005786615302185,0.0023723034026413474,-0.0007144136328255115,0.005453941959808597,0.5161156199227342,0.5118581409595148,0.5205601281866974,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.030713084513084528,0.0280383697983698,0.03368947286572287,0.08937998712998714,0.08109063545688547,0.09836511100386101,14.582967910767913,13.842146657014156,15.261898960210212,0.021675847275847274,0.02076378378378379,0.022689414414414417,0.05835364650364651,0.05572534534534535,0.06092968575718576,17.194992170742168,16.10918240830116,18.36624475707851
1 method pair clip_count draw_count same_time_similarity_video_macro_mean same_time_similarity_ci95_low same_time_similarity_ci95_high video_id_count shifted_far_similarity_video_macro_mean shifted_far_similarity_ci95_low shifted_far_similarity_ci95_high same_minus_shifted_margin_video_macro_mean same_minus_shifted_margin_ci95_low same_minus_shifted_margin_ci95_high matched_vs_shifted_auc_video_macro_mean matched_vs_shifted_auc_ci95_low matched_vs_shifted_auc_ci95_high exact_r1_audio_to_vision_video_macro_mean exact_r1_audio_to_vision_ci95_low exact_r1_audio_to_vision_ci95_high within_pm1_r1_audio_to_vision_video_macro_mean within_pm1_r1_audio_to_vision_ci95_low within_pm1_r1_audio_to_vision_ci95_high mase_slots_audio_to_vision_video_macro_mean mase_slots_audio_to_vision_ci95_low mase_slots_audio_to_vision_ci95_high exact_r1_vision_to_audio_video_macro_mean exact_r1_vision_to_audio_ci95_low exact_r1_vision_to_audio_ci95_high within_pm1_r1_vision_to_audio_video_macro_mean within_pm1_r1_vision_to_audio_ci95_low within_pm1_r1_vision_to_audio_ci95_high mase_slots_vision_to_audio_video_macro_mean mase_slots_vision_to_audio_ci95_low mase_slots_vision_to_audio_ci95_high exact_r1_text_to_audio_video_macro_mean exact_r1_text_to_audio_ci95_low exact_r1_text_to_audio_ci95_high within_pm1_r1_text_to_audio_video_macro_mean within_pm1_r1_text_to_audio_ci95_low within_pm1_r1_text_to_audio_ci95_high mase_slots_text_to_audio_video_macro_mean mase_slots_text_to_audio_ci95_low mase_slots_text_to_audio_ci95_high exact_r1_audio_to_text_video_macro_mean exact_r1_audio_to_text_ci95_low exact_r1_audio_to_text_ci95_high within_pm1_r1_audio_to_text_video_macro_mean within_pm1_r1_audio_to_text_ci95_low within_pm1_r1_audio_to_text_ci95_high mase_slots_audio_to_text_video_macro_mean mase_slots_audio_to_text_ci95_low mase_slots_audio_to_text_ci95_high exact_r1_text_to_vision_video_macro_mean exact_r1_text_to_vision_ci95_low exact_r1_text_to_vision_ci95_high within_pm1_r1_text_to_vision_video_macro_mean within_pm1_r1_text_to_vision_ci95_low within_pm1_r1_text_to_vision_ci95_high mase_slots_text_to_vision_video_macro_mean mase_slots_text_to_vision_ci95_low mase_slots_text_to_vision_ci95_high exact_r1_vision_to_text_video_macro_mean exact_r1_vision_to_text_ci95_low exact_r1_vision_to_text_ci95_high within_pm1_r1_vision_to_text_video_macro_mean within_pm1_r1_vision_to_text_ci95_low within_pm1_r1_vision_to_text_ci95_high mase_slots_vision_to_text_video_macro_mean mase_slots_vision_to_text_ci95_low mase_slots_vision_to_text_ci95_high
2 M3_noSourceTime audio_vision 100 1.0 0.02983147300783216 -0.00780859958706209 0.06532931063779485 37 0.026275558042943677 -0.0135765906084752 0.06442367880532844 0.0035559149673156645 0.0011573716778813968 0.0058081254094310905 0.514271623637315 0.5086281381400399 0.5205522070322601 0.02223723723723724 0.019318307593307594 0.0252463963963964 0.0629030459030459 0.056346739596739595 0.0701120549120549 16.06521836121836 15.360586422136418 16.831551512226515 0.02093822393822394 0.01916058558558559 0.022803474903474903 0.0588974688974689 0.05515479407979409 0.06256855426855427 17.3258314028314 16.454282078507074 18.303939736164736
3 M3_noSourceTime text_audio 100 1.0 0.20791600453897077 0.18457893455935243 0.23068674583726764 37 0.0660212989394217 0.04607872245037598 0.08693206487027433 0.14189470507476618 0.1258566735337873 0.158192415668532 0.7246991238959324 0.7032607291086682 0.7463028065299342 0.13955383955383954 0.12312348777348776 0.156250858000858 0.27616302016302025 0.24729871299871298 0.30469959244959244 10.11119390819391 9.083786154011154 11.220066763191763 0.07386400686400688 0.06408464178464178 0.08439125911625915 0.16222651222651222 0.1441163770913771 0.18078545688545689 12.965773916773918 12.10177589017589 13.855149796224795
4 M3_noSourceTime text_vision 100 1.0 0.032579638831090094 0.012203289357956712 0.053855902535509696 37 0.020862086972797865 -0.0005364448539057191 0.04502733759713527 0.01171755187282541 0.007967536814828653 0.015164253928682914 0.535601545504471 0.5297980337765578 0.5417672611678928 0.04206177606177607 0.03788120978120979 0.04638886743886744 0.10588416988416989 0.09782853925353927 0.113992277992278 14.162812097812097 13.519815894465893 14.831819369369372 0.023044616044616047 0.02111688116688117 0.025180180180180183 0.06238695838695839 0.05710733590733591 0.06806795366795368 17.50512955812956 16.450267953667954 18.578393522093524
5 M3_sourceTime audio_vision 100 1.0 0.007696060061626377 -0.029554258849169186 0.04545826630162894 37 0.008636827414143384 -0.030810690011533682 0.04782087975048893 -0.0009407673630797903 -0.003493066123958016 0.0017008874928592713 0.5026440831028598 0.49047663000196245 0.5166576717342342 0.021732303732303733 0.018566891891891892 0.02489736164736165 0.05924495924495925 0.05102570785070786 0.06822308022308023 18.418811239811237 17.145313781638777 19.668025257400256 0.01908751608751609 0.017070527670527674 0.021123584298584297 0.05492621192621192 0.05029851994851995 0.05956598026598027 18.342253110253115 17.42752445302445 19.27038470613471
6 M3_sourceTime text_audio 100 1.0 0.030779919639574976 0.009495005332440814 0.051172133170129974 37 0.012129069987308123 -0.008134753400298974 0.03220063209724045 0.018650849718729958 0.01209877043793774 0.025570886226139 0.5453406619005025 0.5310778723489828 0.5597143654149971 0.03846546546546547 0.03189687902187902 0.04500438652938652 0.0974937794937795 0.08529543114543116 0.1109557915057915 16.05845045045045 15.243514607464608 16.823768436293438 0.026703560703560704 0.022845398970398972 0.030940111540111538 0.07566495066495067 0.06590900900900902 0.08547635135135136 15.480417417417419 14.34030947018447 16.63520718575719
7 M3_sourceTime text_vision 100 1.0 0.0147220734507245 -0.008115254084484056 0.03707853268159794 37 0.014395209505575931 -0.008901017726582194 0.03943286969561065 0.0003268639452983987 -0.002732993650720293 0.0035072161571967925 0.4998074879590837 0.4950654246809501 0.504675911331164 0.020641784641784643 0.017091269841269845 0.024615486915486918 0.05474774774774775 0.04679173101673103 0.0628289897039897 19.39496525096525 18.298798712998718 20.349694851994855 0.020431574431574433 0.01848766623766624 0.02242575075075075 0.05971514371514372 0.05510990990990991 0.06392915057915058 16.647335478335478 15.535322372372372 17.931235296010293
8 M4_sourceTime audio_vision 100 1.0 0.017589288202760855 -0.021384537387894743 0.05702271457169135 37 0.018760167504660787 -0.01933644305751479 0.05750124747004294 -0.0011708792981242977 -0.004425662602822854 0.0016943904201328112 0.5092526072805329 0.49760200300984875 0.5208348244378497 0.018691977691977692 0.014877273702273703 0.02274350064350065 0.04963020163020163 0.041014339339339345 0.05834750107250107 18.292485628485622 16.987942664092664 19.485725182325183 0.02013041613041613 0.017474710424710424 0.022690025740025738 0.05348219648219649 0.04821073573573574 0.05895694980694981 18.10186143286143 16.83444991419991 19.40507165379665
9 M4_sourceTime text_audio 100 1.0 0.02029622882045845 -0.015886606029959244 0.052645537742046604 37 0.004637936173545868 -0.02838808700016482 0.03639667112261427 0.01565829274056424 0.009190044274264193 0.022804289586498057 0.5389472080363038 0.514210176409312 0.5634661651963513 0.027861861861861865 0.02250713213213213 0.033694744744744744 0.06838953238953241 0.05814036894036893 0.07883789146289145 18.20780223080223 16.355241891891897 20.29683777348777 0.030084513084513088 0.02537934362934363 0.03572420634920635 0.07645002145002144 0.06447318747318749 0.09150033247533246 17.343640926640923 16.01122001287001 18.94879967824968
10 M4_sourceTime text_vision 100 1.0 0.033176119103082845 -0.0041519938906857895 0.07124988025892566 37 0.030836823717937716 -0.008539991982258287 0.0710771330847214 0.0023392953867183023 -0.001506538955824979 0.006041902250845819 0.5036913011035352 0.4923280641013353 0.5162466670602972 0.018344916344916345 0.015335810810810813 0.021540604890604893 0.05032947232947234 0.042268693693693696 0.05863465250965252 18.6358845988846 17.37207688760189 20.017716023166024 0.020106392106392104 0.017946728871728875 0.022438331188331193 0.051333762333762335 0.045313084513084516 0.057213953238953244 19.317567996567995 17.859128196053195 20.795619787644785
11 TSFA-global audio_vision 100 1.0 0.06059821270500538 0.023914981549968812 0.09491150040596706 37 0.03732938954030518 0.0038612152410352285 0.07181022922532529 0.02326882317339013 0.016046146106362763 0.031464541216613816 0.5711938543786947 0.5580918275513356 0.5851307426062079 0.05056928356928358 0.04562037752037753 0.055837269412269415 0.13590690690690693 0.12514061561561562 0.1478566066066066 13.252549978549979 12.300325911625912 14.232603314028314 0.02580952380952381 0.022693564993564997 0.02993452380952381 0.07257400257400258 0.06430289575289577 0.08383320463320464 16.92865637065637 15.890422136422137 17.992037709137712
12 TSFA-global text_audio 100 1.0 0.21948498332128252 0.18447166931635653 0.25264341574445714 37 0.13784915199225622 0.10805201589136158 0.16772503723338508 0.0816358312123013 0.0694722462373281 0.09469795330548046 0.6997278072145092 0.6772065253418312 0.7246179070170095 0.05226340626340627 0.043850450450450454 0.061542020592020574 0.1483732303732304 0.12583120978120982 0.1716539897039897 10.44393564993565 9.248472972972973 11.684178474903476 0.050666237666237665 0.04418186400686401 0.057607464607464604 0.13168897468897467 0.11565889103389104 0.1511234341484341 13.042119691119693 11.967449710424711 14.109006949806952
13 TSFA-global text_vision 100 1.0 0.03882392940236514 -0.005101331978814988 0.08023915360506113 37 0.026369558634260028 -0.019257361349713748 0.0695494351361847 0.012454370779342111 0.006579723199721584 0.018851628858371938 0.5523689941841539 0.5389573468177523 0.5678277126290424 0.036459888459888457 0.032251780351780356 0.040649946374946386 0.10280694980694982 0.09098980051480052 0.11672229729729731 13.241670098670095 12.166959716859719 14.325898702273703 0.02583869583869584 0.02256068211068211 0.030118897468897468 0.06330373230373229 0.05587844272844273 0.07113738738738737 18.464465465465462 17.013464017589012 19.91242631917632
14 TSFA-main audio_vision 100 1.0 0.013561541277063027 -0.01457795284016425 0.04262594169249218 37 0.008980510418424372 -0.021758340514473864 0.04004785757980806 0.004581030828024072 -0.0002248605515836633 0.010127997422592447 0.5114767559162506 0.5016843428677538 0.5212400272042559 0.02182110682110682 0.017747586872586874 0.02623365508365508 0.061913770913770906 0.05110411840411841 0.07329221364221364 17.73442985842986 16.826004118404118 18.77729813384814 0.022406692406692408 0.019272361647361645 0.026145913770913774 0.06514157014157014 0.056082850707850705 0.07448922136422137 15.752643071643071 14.584650504075505 16.82486615186615
15 TSFA-main text_audio 100 1.0 0.11990332463810639 0.0898675046922316 0.15305510010030474 37 0.07596570038282524 0.04989764804945704 0.10443112074815332 0.04393762438637949 0.03607569399696766 0.052024387539309866 0.6394096191556298 0.612673730897501 0.667128455450131 0.043396825396825406 0.03658292578292579 0.05185294937794939 0.12236121836121837 0.10395112612612614 0.14359046546546542 12.342245388245388 11.084742546117546 13.683040540540542 0.039102960102960105 0.03286315958815959 0.04523687258687259 0.10152981552981553 0.08593539253539254 0.11686853281853281 14.241285285285283 13.037042556842557 15.514015894465896
16 TSFA-main text_vision 100 1.0 0.04098847119260702 0.005582397132704427 0.07929293901031172 37 0.037891365960320264 0.0005784316926466229 0.07758778326306248 0.0030971052297696644 -0.000814157517568749 0.0067062238179970675 0.5073206025174111 0.49017726806879075 0.5255564948211889 0.018722436722436724 0.014964897039897044 0.022390036465036463 0.05502488202488203 0.04573553196053196 0.06428104890604891 18.6562737022737 17.207105051480053 20.233926383526384 0.020561561561561564 0.01860630630630631 0.022591741741741746 0.05686486486486487 0.05081441441441443 0.06266833976833977 18.087792363792367 16.792936625911626 19.410561314886316
17 TSFA-multiply audio_vision 100 1.0 0.00978179982461232 -0.02070967490279534 0.0421130031397444 37 0.0060832866483517 -0.02193733706144022 0.03656021604936627 0.003698513197226362 -0.0011122262813486495 0.009267909918290072 0.5129411883387948 0.5027842064281226 0.523200954159897 0.022283998283998283 0.017675825825825827 0.027219133419133415 0.05865551265551267 0.047279225654225654 0.07083616473616473 17.81548777348777 16.808302638352636 18.797270334620336 0.021042900042900044 0.01781427498927499 0.024212129987129988 0.0629056199056199 0.05411741741741742 0.07141601244101244 15.856422565422568 14.7419083976834 16.96319958172458
18 TSFA-multiply text_audio 100 1.0 0.11991994317142747 0.0889362489426478 0.15110231629245585 37 0.07792934440600745 0.052977172014714184 0.10278047049072589 0.04199059878088213 0.034127040037302034 0.05013125747396997 0.6336595358656528 0.6087905367706232 0.6589064419710364 0.04137323037323038 0.034712870012870024 0.048116763191763195 0.11405105105105105 0.09659961389961391 0.13316510081510083 12.645648219648223 11.431381585156586 13.97510311025311 0.04315615615615616 0.03623397683397684 0.05003088803088803 0.1083843843843844 0.09482691977691977 0.12210212355212356 14.16602059202059 13.052271911196911 15.273814092664091
19 TSFA-multiply text_vision 100 1.0 0.03612421115527528 0.003679617927233912 0.0706390626773947 37 0.033054201762203535 0.0003160863252284142 0.06742515595895551 0.0030700093951693155 -0.0007710019667443696 0.006867931876338059 0.5052031858986646 0.4900221838004418 0.5219854911997864 0.01671685971685972 0.013541377091377095 0.019720752895752897 0.04806392106392106 0.03873852423852424 0.05764617117117116 19.210052767052765 17.75446628056628 20.751913610038613 0.021712569712569713 0.019381831831831835 0.02410126555126555 0.05798712998712999 0.05170907335907336 0.06397966537966537 18.30359202059202 16.94032631917632 19.627432818532814
20 TSFA-random audio_vision 100 20.0 0.020861894386615707 0.005693970997846327 0.037061795072022896 37 0.01680333015403859 0.0004332554102972204 0.03303836082632705 0.0040585642224212435 0.0005754381947093255 0.007718638619050609 0.5193708956724581 0.5150352463917897 0.5238890623406223 0.03257623337623338 0.030417548262548257 0.0346354971042471 0.09581681681681682 0.08960499839124837 0.10207584298584299 15.154294980694983 14.546768505469755 15.738201547619045 0.021379450879450878 0.02055578024453025 0.022163257722007723 0.06266823251823253 0.060993607357357364 0.0645759614972115 16.373181810381816 15.610742067245818 17.204955935220937
21 TSFA-random text_audio 100 20.0 0.17050077302726946 0.13941037026448067 0.1990060504478285 37 0.09672763162535135 0.07223268336231446 0.12007034012783693 0.07377314134437793 0.06320199140044352 0.08493496876685644 0.6780049539566163 0.6551398515962296 0.7010971646109445 0.048387923637923644 0.04305602799227799 0.05432789575289575 0.13999991419991417 0.12458882453882453 0.15700594272844276 11.147006499356502 10.136636986808236 12.213333453989703 0.04792293007293008 0.04362501340626341 0.051922379343629345 0.12110746460746463 0.11033633526383527 0.13127376447876446 13.276559244959246 12.25213515122265 14.3009197506435
22 TSFA-random text_vision 100 20.0 0.028480983388760747 -0.0010045256113203147 0.05799119457573753 37 0.026108679984648473 -0.004959284461738992 0.055005786615302185 0.0023723034026413474 -0.0007144136328255115 0.005453941959808597 0.5161156199227342 0.5118581409595148 0.5205601281866974 0.030713084513084528 0.0280383697983698 0.03368947286572287 0.08937998712998714 0.08109063545688547 0.09836511100386101 14.582967910767913 13.842146657014156 15.261898960210212 0.021675847275847274 0.02076378378378379 0.022689414414414417 0.05835364650364651 0.05572534534534535 0.06092968575718576 17.194992170742168 16.10918240830116 18.36624475707851
@@ -0,0 +1,11 @@
left_method,right_method,metric,clip_count,video_id_count,left_minus_right_video_macro_mean,ci95_low,ci95_high
TSFA-main,M4_sourceTime,content_auc_mean,100,37,0.035438620389639894,0.023293646690953858,0.04698819770251506
TSFA-main,M4_sourceTime,canonical_pairwise_time_mae_mean,100,37,0.005933525736358895,0.005296561964374039,0.006509589017113389
TSFA-main,TSFA-random,content_auc_mean,100,37,-0.01842816398750574,-0.02718789359398819,-0.008992237898280122
TSFA-main,TSFA-random,canonical_pairwise_time_mae_mean,100,37,-0.18037076895618454,-0.18518644583548508,-0.17535325008436822
TSFA-main,TSFA-global,content_auc_mean,100,37,-0.055027892729355495,-0.06412401690967184,-0.04722633193128541
TSFA-main,TSFA-global,canonical_pairwise_time_mae_mean,100,37,-0.17289980508939898,-0.17827689029341662,-0.16739238598647427
TSFA-multiply,TSFA-main,content_auc_mean,100,37,-0.0021343558287264036,-0.003982915698119604,-0.00020908619168815494
TSFA-multiply,TSFA-main,canonical_pairwise_time_mae_mean,100,37,-0.0025367852317495707,-0.002737516713424809,-0.002348100112530585
TSFA-main,M3_noSourceTime,content_auc_mean,100,37,-0.038788438482809065,-0.05495972110741062,-0.02010023840091993
TSFA-main,M3_noSourceTime,canonical_pairwise_time_mae_mean,100,37,-0.18092685065902986,-0.18653708198539262,-0.17516024678207237
1 left_method right_method metric clip_count video_id_count left_minus_right_video_macro_mean ci95_low ci95_high
2 TSFA-main M4_sourceTime content_auc_mean 100 37 0.035438620389639894 0.023293646690953858 0.04698819770251506
3 TSFA-main M4_sourceTime canonical_pairwise_time_mae_mean 100 37 0.005933525736358895 0.005296561964374039 0.006509589017113389
4 TSFA-main TSFA-random content_auc_mean 100 37 -0.01842816398750574 -0.02718789359398819 -0.008992237898280122
5 TSFA-main TSFA-random canonical_pairwise_time_mae_mean 100 37 -0.18037076895618454 -0.18518644583548508 -0.17535325008436822
6 TSFA-main TSFA-global content_auc_mean 100 37 -0.055027892729355495 -0.06412401690967184 -0.04722633193128541
7 TSFA-main TSFA-global canonical_pairwise_time_mae_mean 100 37 -0.17289980508939898 -0.17827689029341662 -0.16739238598647427
8 TSFA-multiply TSFA-main content_auc_mean 100 37 -0.0021343558287264036 -0.003982915698119604 -0.00020908619168815494
9 TSFA-multiply TSFA-main canonical_pairwise_time_mae_mean 100 37 -0.0025367852317495707 -0.002737516713424809 -0.002348100112530585
10 TSFA-main M3_noSourceTime content_auc_mean 100 37 -0.038788438482809065 -0.05495972110741062 -0.02010023840091993
11 TSFA-main M3_noSourceTime canonical_pairwise_time_mae_mean 100 37 -0.18092685065902986 -0.18653708198539262 -0.17516024678207237
@@ -0,0 +1,17 @@
{
"created_utc": "2026-09-23T14:32:44.179041+00:00",
"experiment": "TSFA shared raw-content alignment-only probe",
"sample_count": 100,
"heldout_count": 100,
"video_id_count": 37,
"fold_count": 5,
"seed": 42,
"delta": 0.1,
"probe_epochs": 40,
"random_window_repeats": 20,
"feature_protocol": "Within each fold, feature normalization is fitted on 80 training clips. For every method and modality, the same normalized raw source feature matrix is pooled using that method's A^m; native M3/M4/TSFA value and output projections are excluded from scored features.",
"probe_protocol": "One 64-dimensional linear projector per modality, trained on 80 training clips with the same within-clip InfoNCE protocol and identical fold seed across methods. Scores use only 20 held-out clips per fold.",
"interpretation_limit": "A same-slot positive is a timestamp/slot convention, not independently annotated semantic ground truth. Source-time attention can still encode time through selected raw values.",
"device": "cuda",
"elapsed_seconds": 63.806029081344604
}
@@ -0,0 +1,9 @@
method,modality,clip_count,video_id_count,candidate_count_mean_video_macro_mean,candidate_count_mean_ci95_low,candidate_count_mean_ci95_high,candidate_fraction_valid_mean_video_macro_mean,candidate_fraction_valid_mean_ci95_low,candidate_fraction_valid_mean_ci95_high,temporal_prior_mass_mean_video_macro_mean,temporal_prior_mass_mean_ci95_low,temporal_prior_mass_mean_ci95_high,candidate_mean_center_error_video_macro_mean,candidate_mean_center_error_ci95_low,candidate_mean_center_error_ci95_high
TSFA-global,audio,100,37,859.2909266409266,725.903935006435,1010.8943355855854,0.9999999840133748,0.9999999739948848,0.9999999929042089,1.0,1.0,1.0,0.23349895292516687,0.23334358473436637,0.23365915023423298
TSFA-global,vision,100,37,43.82829257829257,37.07407764907765,51.22568103818104,0.9999999899584969,0.9999999802928788,0.9999999976372933,1.0,1.0,1.0,0.23209808292734374,0.23138750004937297,0.23281616748889555
TSFA-main,audio,100,37,173.57981845948993,148.11770421039185,204.1688543426484,0.2027066824698029,0.20226793367661783,0.20315699026129216,0.7372228195819668,0.7367319915047317,0.7377194487338015,0.0008611694746814316,0.0007925661735736586,0.0009352763757801223
TSFA-main,vision,100,37,8.697594809082126,7.356396586381408,10.302393176907227,0.19806601293752601,0.19761108234273017,0.19851417924334616,0.7214869179295039,0.7187354109908368,0.7239951614850468,0.005114482842239538,0.004486957946824082,0.0057506511149799265
TSFA-multiply,audio,100,37,173.57981845948993,147.94183613935493,206.29472111888524,0.2027066824698029,0.20227122749140025,0.20313434986607776,0.7372228195819668,0.7367260256056134,0.737726254887409,0.0008611694746814316,0.0007943149040772448,0.0009299756654477927
TSFA-multiply,vision,100,37,8.697594809082126,7.429751027647465,10.313039470908723,0.19806601293752601,0.19755518793732132,0.19851298628820818,0.7214869179295039,0.7185624315606288,0.7239946865184925,0.005114482842239538,0.004483692884955575,0.0057248699215185035
TSFA-random,audio,100,37,165.4303607225725,140.97407569611394,193.95208463450228,0.1933617310581552,0.19281956570094802,0.1938787973451431,0.19230555606103,0.19073511667908855,0.19371052016348292,0.005423930897291032,0.00533451006593477,0.0055040052315254925
TSFA-random,vision,100,37,8.36343770343799,7.137826227282222,9.895227294917355,0.19097706449311183,0.1904804513221966,0.19147352101705237,0.19106015145038713,0.18928042240004053,0.1929521264620367,0.007972950431270442,0.007579205200452694,0.008422631476527104
1 method modality clip_count video_id_count candidate_count_mean_video_macro_mean candidate_count_mean_ci95_low candidate_count_mean_ci95_high candidate_fraction_valid_mean_video_macro_mean candidate_fraction_valid_mean_ci95_low candidate_fraction_valid_mean_ci95_high temporal_prior_mass_mean_video_macro_mean temporal_prior_mass_mean_ci95_low temporal_prior_mass_mean_ci95_high candidate_mean_center_error_video_macro_mean candidate_mean_center_error_ci95_low candidate_mean_center_error_ci95_high
2 TSFA-global audio 100 37 859.2909266409266 725.903935006435 1010.8943355855854 0.9999999840133748 0.9999999739948848 0.9999999929042089 1.0 1.0 1.0 0.23349895292516687 0.23334358473436637 0.23365915023423298
3 TSFA-global vision 100 37 43.82829257829257 37.07407764907765 51.22568103818104 0.9999999899584969 0.9999999802928788 0.9999999976372933 1.0 1.0 1.0 0.23209808292734374 0.23138750004937297 0.23281616748889555
4 TSFA-main audio 100 37 173.57981845948993 148.11770421039185 204.1688543426484 0.2027066824698029 0.20226793367661783 0.20315699026129216 0.7372228195819668 0.7367319915047317 0.7377194487338015 0.0008611694746814316 0.0007925661735736586 0.0009352763757801223
5 TSFA-main vision 100 37 8.697594809082126 7.356396586381408 10.302393176907227 0.19806601293752601 0.19761108234273017 0.19851417924334616 0.7214869179295039 0.7187354109908368 0.7239951614850468 0.005114482842239538 0.004486957946824082 0.0057506511149799265
6 TSFA-multiply audio 100 37 173.57981845948993 147.94183613935493 206.29472111888524 0.2027066824698029 0.20227122749140025 0.20313434986607776 0.7372228195819668 0.7367260256056134 0.737726254887409 0.0008611694746814316 0.0007943149040772448 0.0009299756654477927
7 TSFA-multiply vision 100 37 8.697594809082126 7.429751027647465 10.313039470908723 0.19806601293752601 0.19755518793732132 0.19851298628820818 0.7214869179295039 0.7185624315606288 0.7239946865184925 0.005114482842239538 0.004483692884955575 0.0057248699215185035
8 TSFA-random audio 100 37 165.4303607225725 140.97407569611394 193.95208463450228 0.1933617310581552 0.19281956570094802 0.1938787973451431 0.19230555606103 0.19073511667908855 0.19371052016348292 0.005423930897291032 0.00533451006593477 0.0055040052315254925
9 TSFA-random vision 100 37 8.36343770343799 7.137826227282222 9.895227294917355 0.19097706449311183 0.1904804513221966 0.19147352101705237 0.19106015145038713 0.18928042240004053 0.1929521264620367 0.007972950431270442 0.007579205200452694 0.008422631476527104
@@ -0,0 +1,22 @@
method,pair,clip_count,draw_count,same_time_similarity_video_macro_mean,same_time_similarity_ci95_low,same_time_similarity_ci95_high,video_id_count,shifted_far_similarity_video_macro_mean,shifted_far_similarity_ci95_low,shifted_far_similarity_ci95_high,same_minus_shifted_margin_video_macro_mean,same_minus_shifted_margin_ci95_low,same_minus_shifted_margin_ci95_high,matched_vs_shifted_auc_video_macro_mean,matched_vs_shifted_auc_ci95_low,matched_vs_shifted_auc_ci95_high,exact_r1_audio_to_vision_video_macro_mean,exact_r1_audio_to_vision_ci95_low,exact_r1_audio_to_vision_ci95_high,within_pm1_r1_audio_to_vision_video_macro_mean,within_pm1_r1_audio_to_vision_ci95_low,within_pm1_r1_audio_to_vision_ci95_high,mase_slots_audio_to_vision_video_macro_mean,mase_slots_audio_to_vision_ci95_low,mase_slots_audio_to_vision_ci95_high,exact_r1_vision_to_audio_video_macro_mean,exact_r1_vision_to_audio_ci95_low,exact_r1_vision_to_audio_ci95_high,within_pm1_r1_vision_to_audio_video_macro_mean,within_pm1_r1_vision_to_audio_ci95_low,within_pm1_r1_vision_to_audio_ci95_high,mase_slots_vision_to_audio_video_macro_mean,mase_slots_vision_to_audio_ci95_low,mase_slots_vision_to_audio_ci95_high,exact_r1_text_to_audio_video_macro_mean,exact_r1_text_to_audio_ci95_low,exact_r1_text_to_audio_ci95_high,within_pm1_r1_text_to_audio_video_macro_mean,within_pm1_r1_text_to_audio_ci95_low,within_pm1_r1_text_to_audio_ci95_high,mase_slots_text_to_audio_video_macro_mean,mase_slots_text_to_audio_ci95_low,mase_slots_text_to_audio_ci95_high,exact_r1_audio_to_text_video_macro_mean,exact_r1_audio_to_text_ci95_low,exact_r1_audio_to_text_ci95_high,within_pm1_r1_audio_to_text_video_macro_mean,within_pm1_r1_audio_to_text_ci95_low,within_pm1_r1_audio_to_text_ci95_high,mase_slots_audio_to_text_video_macro_mean,mase_slots_audio_to_text_ci95_low,mase_slots_audio_to_text_ci95_high,exact_r1_text_to_vision_video_macro_mean,exact_r1_text_to_vision_ci95_low,exact_r1_text_to_vision_ci95_high,within_pm1_r1_text_to_vision_video_macro_mean,within_pm1_r1_text_to_vision_ci95_low,within_pm1_r1_text_to_vision_ci95_high,mase_slots_text_to_vision_video_macro_mean,mase_slots_text_to_vision_ci95_low,mase_slots_text_to_vision_ci95_high,exact_r1_vision_to_text_video_macro_mean,exact_r1_vision_to_text_ci95_low,exact_r1_vision_to_text_ci95_high,within_pm1_r1_vision_to_text_video_macro_mean,within_pm1_r1_vision_to_text_ci95_low,within_pm1_r1_vision_to_text_ci95_high,mase_slots_vision_to_text_video_macro_mean,mase_slots_vision_to_text_ci95_low,mase_slots_vision_to_text_ci95_high
M3_noSourceTime,audio_vision,100,1.0,0.036963944587425426,0.012654210555109019,0.06213064156953601,37,0.01284708399055921,-0.010789174090569271,0.03801324850198797,0.02411686070606988,0.018906500978365978,0.02932537980979977,0.5615005482306014,0.5492661873214401,0.5739596572161066,0.0547975117975118,0.04492898970398971,0.06514066924066923,0.128025311025311,0.11184757614757616,0.1455527027027027,13.629726297726297,12.843761196911194,14.369294884169886,0.03019090519090519,0.026839886314886315,0.033779096954096956,0.07865679965679967,0.07125554483054485,0.08641603389103389,16.362507507507505,15.478168918918918,17.330904729729728,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_noSourceTime,text_audio,100,1.0,0.5428216995771276,0.5157750271393224,0.5659249134957856,37,0.1862772215156948,0.16884496117041078,0.2044323245350392,0.35654447643267695,0.32902113457284277,0.38482697127963067,0.9559461184208523,0.9410152465582983,0.968752133678131,,,,,,,,,,,,,,,,,,,0.36576791076791076,0.3160490561990562,0.421391108966109,0.6394950664950665,0.5802250858000858,0.6932264800514802,2.3383740883740884,1.8257137387387388,2.967019691119691,0.3325842985842986,0.2862534749034749,0.3864762226512226,0.5848228228228229,0.5289179107679108,0.6410546332046333,3.5629060489060485,2.914686679536679,4.190957496782497,,,,,,,,,,,,,,,,,,
M3_noSourceTime,text_vision,100,1.0,0.07063237097845877,0.049190198520261096,0.09198749123186528,37,0.022404568813529854,0.0027337225595416618,0.042923773043187015,0.04822780188845877,0.039832519238572194,0.05659220698571646,0.6063276332943885,0.5904511958719739,0.6233192521834144,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.11063191763191764,0.09528938223938226,0.12811927284427282,0.22938867438867436,0.20507918275418277,0.2533813384813385,10.313163449163447,9.49544851994852,11.128362998712998,0.03617932217932218,0.030936861861861866,0.041976029601029606,0.09173659373659374,0.08058330115830119,0.10378899613899614,15.593415701415703,14.841660489060489,16.365030341055338
M3_sourceTime,audio_vision,100,1.0,-0.01390508033140634,-0.05067550643888827,0.021212040213628297,37,-0.010913303125673809,-0.05113494249747865,0.028314814638195643,-0.002991777190843502,-0.005994554468138316,2.8365758378620055e-05,0.5029486009565798,0.49220838989635535,0.514119726033024,0.02227713427713428,0.01772549335049335,0.027349259974259972,0.05884513084513086,0.04750401115401116,0.07049715787215788,18.302241956241957,17.049701190476192,19.5555055984556,0.020694122694122692,0.01873700128700129,0.022657818532818536,0.06109223509223509,0.05726682754182755,0.06464848777348778,16.701135564135562,15.928820785070785,17.462392299442303,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_sourceTime,text_audio,100,1.0,0.13965087270742027,0.11086805911581128,0.17086291475965698,37,0.07104170450923547,0.04617817302421454,0.09790063937091513,0.06860916809770604,0.05853303224689769,0.07936001767631712,0.6837595762555869,0.6552860960409499,0.712458437802354,,,,,,,,,,,,,,,,,,,0.07891848991848992,0.06314130201630203,0.09810974903474903,0.17487301587301587,0.14614791934791935,0.2047123552123552,11.77520634920635,10.68862012012012,12.995895302445302,0.055659373659373665,0.04583983268983269,0.06627339124839127,0.13317975117975117,0.11395983483483482,0.15161607679107675,12.579315744315746,11.378763942513944,13.799107142857137,,,,,,,,,,,,,,,,,,
M3_sourceTime,text_vision,100,1.0,0.018218856013835664,-0.0019383159056188255,0.03824574691640572,37,0.01843536877672055,-0.0031358283705724124,0.0398007043003149,-0.00021651274608931568,-0.0033959447400838593,0.0026867472750431616,0.5073603928776801,0.5013866661304163,0.5138052809268234,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.02189274989274989,0.017497951522951527,0.026569176319176314,0.06272629772629773,0.053270045045045046,0.07303394465894468,18.673179751179752,17.663443983268984,19.694041184041183,0.01792406692406693,0.015639414414414417,0.02020035392535392,0.052900900900900896,0.04713513513513515,0.058622297297297295,17.553818961818966,16.37964121621622,18.782360049335054
M4_sourceTime,audio_vision,100,1.0,0.00960363132716526,-0.026355777083579862,0.048860038747030224,37,0.008486620442388824,-0.028060019295580236,0.04573340321810003,0.0011170108847764362,-0.0023584936828478765,0.004587398100880476,0.5134115260777495,0.4984700587080242,0.529333335144605,0.02358944658944659,0.01981911196911197,0.027432882882882884,0.06169626769626772,0.05163449163449164,0.07249851994851997,17.65543586443586,16.274862580437578,19.030651254826253,0.0226001716001716,0.019407979407979413,0.026030244530244533,0.06180137280137281,0.05271195838695839,0.07151898326898326,17.76484513084513,16.36237923637924,19.1801465036465,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M4_sourceTime,text_audio,100,1.0,0.06588486519170021,0.03945950347046453,0.09289460044348818,37,0.042297942433741985,0.019241170794713554,0.06466475258021885,0.02358692279788703,0.01496262407228283,0.032121939048117505,0.6160321730354975,0.5843945260158524,0.6481647989887683,,,,,,,,,,,,,,,,,,,0.037331617331617335,0.03060465465465466,0.043963170313170305,0.10766709566709566,0.09256888674388675,0.12209738309738306,13.641798798798801,12.305344755469758,15.062401372801373,0.03636851136851138,0.027877606177606182,0.04520361432861432,0.10497468897468898,0.08676978764478764,0.12320139425139427,13.546082368082368,12.237779311454313,14.863573423423423,,,,,,,,,,,,,,,,,,
M4_sourceTime,text_vision,100,1.0,0.002750558023057728,-0.02300323787854146,0.028440979736176288,37,-0.0017949415156420013,-0.03015163323847945,0.026173505636594735,0.004545499502936101,0.00029588325187894103,0.009067848319586198,0.5036754941681804,0.4941306089880723,0.5134271535022865,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.024956670956670958,0.020241580866580868,0.030214564564564565,0.06877262977262978,0.056675429000429,0.08342528957528957,17.827770913770916,16.393623906048905,19.236555115830114,0.018706134706134706,0.01641861861861862,0.020702370227370228,0.0541947661947662,0.049403925353925356,0.05915512655512656,18.49487387387387,17.468441087516084,19.530955534105537
TSFA-global,audio_vision,100,1.0,0.058836026214662884,0.009433591336071618,0.10898509821607989,37,0.0299474197815315,-0.019026566930953497,0.07779842395164258,0.02888860643662733,0.019019952453169467,0.039115650392871,0.5811011281037877,0.5633139432698278,0.5997097862395071,0.05360531960531961,0.04721345988845988,0.06024396181896182,0.1473792363792364,0.1310375482625483,0.16456441441441436,12.399526383526384,11.223109202059202,13.630522844272843,0.026169884169884174,0.023153153153153153,0.02972972972972973,0.072981552981553,0.06513077005577005,0.0819059738309738,16.35457228657229,15.289285274560278,17.390430094380097,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-global,text_audio,100,1.0,0.6151585970459093,0.592303545010855,0.6363802687550169,37,0.3307069785888169,0.30723476551075196,0.3557131022926266,0.28445161780264955,0.2568491350423748,0.313431063751506,0.888037095225925,0.8713038621448196,0.9025196944665029,,,,,,,,,,,,,,,,,,,0.1913303303303303,0.17120793650793648,0.21155896610896607,0.44082539682539684,0.4059675246675246,0.4759347168597168,4.0553886743886745,3.421013191763192,4.85362648005148,0.13350965250965252,0.11927451737451739,0.1480651222651223,0.3463878163878164,0.31618583226083224,0.37751728871728873,6.496407550407551,5.68826003861004,7.341266816816816,,,,,,,,,,,,,,,,,,
TSFA-global,text_vision,100,1.0,0.05193602225629566,0.00557772744820211,0.09913521469205716,37,0.018522642509243077,-0.031060642072022387,0.06214130753726415,0.03341337973851245,0.023362873667806743,0.044610356791469324,0.5860619295207061,0.5682458135804879,0.6058382392698549,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.0624924924924925,0.05556150793650794,0.0695357679107679,0.1712166452166452,0.15426744959244962,0.18985087945087942,10.319833118833117,9.234121578721577,11.574682368082371,0.027978978978978986,0.02543226083226084,0.030679751179751174,0.08042342342342342,0.07197972972972973,0.08928452380952381,16.16939510939511,14.841324195624193,17.51072467824968
TSFA-main,audio_vision,100,1.0,0.014167320872974634,-0.010923848048597458,0.039202572867964074,37,0.00572396734721069,-0.022427993542694716,0.03323088554438458,0.008443353525896292,0.0026837436866675632,0.015297084395711374,0.5168712160108436,0.5063480574752184,0.5274386673004691,0.02177691977691978,0.01702814242814243,0.026475278850278845,0.059097383097383105,0.0487433075933076,0.06956501501501502,17.385808665808668,16.533373498498495,18.1887395002145,0.020783783783783785,0.017991763191763196,0.02361460746460746,0.059234234234234226,0.0514413770913771,0.06677416344916343,17.36473530673531,16.14145280995281,18.719550675675674,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-main,text_audio,100,1.0,0.44300997549055066,0.41801447102011324,0.4684682627939124,37,0.23093871736075175,0.2083505803346634,0.254020971881992,0.21207125773569485,0.18952878732104791,0.23568765088817542,0.8444317785680818,0.8258805004947956,0.8617723152745259,,,,,,,,,,,,,,,,,,,0.12499742599742601,0.11379501287001287,0.13663513513513512,0.3075298155298155,0.2829487880737881,0.3312437258687259,6.916232089232089,6.001308118833121,7.864674903474903,0.1002827112827113,0.08834907764907765,0.11306663449163447,0.2701518661518662,0.24654092664092664,0.29654121621621626,7.918441441441442,6.990601801801802,8.871822072072073,,,,,,,,,,,,,,,,,,
TSFA-main,text_vision,100,1.0,0.0174827435109615,-0.014795200237153264,0.0498013264192671,37,0.007209132154900075,-0.027171939497727737,0.040232186743695476,0.010273611322660614,0.0044551992623352394,0.01675656997344102,0.5184451704744258,0.5070176356542713,0.5300536991588852,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.02143886743886744,0.017653732303732305,0.02514813384813385,0.06253453453453454,0.05233642213642214,0.07386306306306306,17.168172458172457,16.094917084942086,18.22205849420849,0.021965250965250968,0.019243018018018018,0.024749935649935648,0.06071342771342771,0.05453082368082369,0.0671651866151866,16.860575718575717,15.960335660660665,17.848634663234666
TSFA-multiply,audio_vision,100,1.0,0.011799700871433188,-0.013416462369410848,0.03714225595848734,37,0.003975713585776283,-0.023566553326942314,0.029399206363613498,0.007823987230021269,0.002012223205404298,0.014442908524818563,0.5172232899693006,0.5074870466258432,0.5276387511371553,0.020090519090519093,0.01639286786786787,0.023955534105534106,0.05755426855426856,0.04768575718575719,0.06837712355212355,17.54742471042471,16.713855855855854,18.3979314993565,0.020211497211497214,0.017404579579579582,0.022875203775203776,0.058004719004719,0.05067953667953669,0.06470415057915058,16.951271986271987,15.84848443800944,18.024721492921493,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-multiply,text_audio,100,1.0,0.43704115381661734,0.4115881279842063,0.46182952157855794,37,0.22681150353265547,0.20526582196287743,0.2506810446160017,0.21022965143079514,0.186519824604155,0.23452360996737565,0.8422674450666473,0.8232177948179944,0.8604417612806575,,,,,,,,,,,,,,,,,,,0.12100386100386103,0.10705222007722007,0.13455019305019308,0.3025752895752896,0.27550450450450453,0.3296434684684685,7.047516087516087,6.080864478764479,8.009039800514799,0.10672972972972973,0.09416164736164735,0.11945048262548263,0.2784924924924925,0.2542594487344487,0.3048471042471042,7.982056199056198,7.081995849420849,8.870854268554268,,,,,,,,,,,,,,,,,,
TSFA-multiply,text_vision,100,1.0,0.019027262118925484,-0.011798013240855429,0.05015302798114453,37,0.008701149625835541,-0.025909994276746277,0.04330936423482336,0.010326112528868553,0.004384441093969887,0.016339780587871982,0.5203762987911925,0.5091359831811294,0.5319951762695445,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.021473616473616473,0.017847393822393823,0.025207303732303732,0.06058687258687259,0.05200009652509653,0.06891801801801802,17.13536293436293,16.116098712998713,18.12695836550836,0.022200772200772205,0.01949655727155727,0.02511743886743887,0.05943286143286144,0.053154193479193486,0.0655993993993994,16.844067353067352,15.867397082797082,17.840103292578295
TSFA-random,audio_vision,100,20.0,0.0003412659092388615,-0.012525598498225196,0.01250697673623247,37,-0.002757302584965167,-0.017743520338419794,0.011022231272467209,0.0030985684875587003,0.0007254310763673855,0.005789096312349675,0.520877531278429,0.5157918074214983,0.5265877221121911,0.03191679536679536,0.0298980518018018,0.0340041924066924,0.09458826683826685,0.08867857142857143,0.10115269626769628,15.031224281424278,14.405575359287857,15.620949209030458,0.021313685113685112,0.02049507185757186,0.022244750643500646,0.06297726297726298,0.06123326576576577,0.0648488315100815,16.687722222222224,15.855059311990566,17.524219135027884,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-random,text_audio,100,20.0,0.5176953808174132,0.49329636456565923,0.5421801408262926,37,0.27252678952364234,0.24728572903165366,0.2993041891233309,0.24516859096668722,0.22255121880186324,0.26788503176686185,0.8510175559155282,0.8342504583359005,0.8671632742516454,,,,,,,,,,,,,,,,,,,0.13294581724581728,0.12157137548262546,0.14623681306306305,0.35170619905619904,0.32670355051480054,0.37593875482625483,5.308989296439297,4.696646238202488,5.932726235521235,0.11373277563277563,0.1038839065851566,0.12520446160446164,0.2932850278850279,0.2702927622265122,0.3166170704633205,7.3795730802230795,6.716822666237666,8.059022419026167,,,,,,,,,,,,,,,,,,
TSFA-random,text_vision,100,20.0,0.01558142443491039,-0.004607028207164052,0.0340684453611749,37,0.01131799393027936,-0.01173588358127808,0.03077004442264562,0.0042634304992695575,0.0006321330527004355,0.008231108258869975,0.523311412590402,0.5182214701491481,0.5286413477330866,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.03621218361218361,0.03320817889317888,0.03949472597597597,0.1053566495066495,0.09686440047190047,0.11374875321750319,14.090841570141572,13.219035881595882,14.95690251287001,0.021406113256113257,0.02049555823680824,0.022317504826254827,0.06158863148863149,0.05928467771342772,0.06384178035178034,16.919727498927497,15.876626134706132,18.00174117599742
1 method pair clip_count draw_count same_time_similarity_video_macro_mean same_time_similarity_ci95_low same_time_similarity_ci95_high video_id_count shifted_far_similarity_video_macro_mean shifted_far_similarity_ci95_low shifted_far_similarity_ci95_high same_minus_shifted_margin_video_macro_mean same_minus_shifted_margin_ci95_low same_minus_shifted_margin_ci95_high matched_vs_shifted_auc_video_macro_mean matched_vs_shifted_auc_ci95_low matched_vs_shifted_auc_ci95_high exact_r1_audio_to_vision_video_macro_mean exact_r1_audio_to_vision_ci95_low exact_r1_audio_to_vision_ci95_high within_pm1_r1_audio_to_vision_video_macro_mean within_pm1_r1_audio_to_vision_ci95_low within_pm1_r1_audio_to_vision_ci95_high mase_slots_audio_to_vision_video_macro_mean mase_slots_audio_to_vision_ci95_low mase_slots_audio_to_vision_ci95_high exact_r1_vision_to_audio_video_macro_mean exact_r1_vision_to_audio_ci95_low exact_r1_vision_to_audio_ci95_high within_pm1_r1_vision_to_audio_video_macro_mean within_pm1_r1_vision_to_audio_ci95_low within_pm1_r1_vision_to_audio_ci95_high mase_slots_vision_to_audio_video_macro_mean mase_slots_vision_to_audio_ci95_low mase_slots_vision_to_audio_ci95_high exact_r1_text_to_audio_video_macro_mean exact_r1_text_to_audio_ci95_low exact_r1_text_to_audio_ci95_high within_pm1_r1_text_to_audio_video_macro_mean within_pm1_r1_text_to_audio_ci95_low within_pm1_r1_text_to_audio_ci95_high mase_slots_text_to_audio_video_macro_mean mase_slots_text_to_audio_ci95_low mase_slots_text_to_audio_ci95_high exact_r1_audio_to_text_video_macro_mean exact_r1_audio_to_text_ci95_low exact_r1_audio_to_text_ci95_high within_pm1_r1_audio_to_text_video_macro_mean within_pm1_r1_audio_to_text_ci95_low within_pm1_r1_audio_to_text_ci95_high mase_slots_audio_to_text_video_macro_mean mase_slots_audio_to_text_ci95_low mase_slots_audio_to_text_ci95_high exact_r1_text_to_vision_video_macro_mean exact_r1_text_to_vision_ci95_low exact_r1_text_to_vision_ci95_high within_pm1_r1_text_to_vision_video_macro_mean within_pm1_r1_text_to_vision_ci95_low within_pm1_r1_text_to_vision_ci95_high mase_slots_text_to_vision_video_macro_mean mase_slots_text_to_vision_ci95_low mase_slots_text_to_vision_ci95_high exact_r1_vision_to_text_video_macro_mean exact_r1_vision_to_text_ci95_low exact_r1_vision_to_text_ci95_high within_pm1_r1_vision_to_text_video_macro_mean within_pm1_r1_vision_to_text_ci95_low within_pm1_r1_vision_to_text_ci95_high mase_slots_vision_to_text_video_macro_mean mase_slots_vision_to_text_ci95_low mase_slots_vision_to_text_ci95_high
2 M3_noSourceTime audio_vision 100 1.0 0.036963944587425426 0.012654210555109019 0.06213064156953601 37 0.01284708399055921 -0.010789174090569271 0.03801324850198797 0.02411686070606988 0.018906500978365978 0.02932537980979977 0.5615005482306014 0.5492661873214401 0.5739596572161066 0.0547975117975118 0.04492898970398971 0.06514066924066923 0.128025311025311 0.11184757614757616 0.1455527027027027 13.629726297726297 12.843761196911194 14.369294884169886 0.03019090519090519 0.026839886314886315 0.033779096954096956 0.07865679965679967 0.07125554483054485 0.08641603389103389 16.362507507507505 15.478168918918918 17.330904729729728
3 M3_noSourceTime text_audio 100 1.0 0.5428216995771276 0.5157750271393224 0.5659249134957856 37 0.1862772215156948 0.16884496117041078 0.2044323245350392 0.35654447643267695 0.32902113457284277 0.38482697127963067 0.9559461184208523 0.9410152465582983 0.968752133678131 0.36576791076791076 0.3160490561990562 0.421391108966109 0.6394950664950665 0.5802250858000858 0.6932264800514802 2.3383740883740884 1.8257137387387388 2.967019691119691 0.3325842985842986 0.2862534749034749 0.3864762226512226 0.5848228228228229 0.5289179107679108 0.6410546332046333 3.5629060489060485 2.914686679536679 4.190957496782497
4 M3_noSourceTime text_vision 100 1.0 0.07063237097845877 0.049190198520261096 0.09198749123186528 37 0.022404568813529854 0.0027337225595416618 0.042923773043187015 0.04822780188845877 0.039832519238572194 0.05659220698571646 0.6063276332943885 0.5904511958719739 0.6233192521834144 0.11063191763191764 0.09528938223938226 0.12811927284427282 0.22938867438867436 0.20507918275418277 0.2533813384813385 10.313163449163447 9.49544851994852 11.128362998712998 0.03617932217932218 0.030936861861861866 0.041976029601029606 0.09173659373659374 0.08058330115830119 0.10378899613899614 15.593415701415703 14.841660489060489 16.365030341055338
5 M3_sourceTime audio_vision 100 1.0 -0.01390508033140634 -0.05067550643888827 0.021212040213628297 37 -0.010913303125673809 -0.05113494249747865 0.028314814638195643 -0.002991777190843502 -0.005994554468138316 2.8365758378620055e-05 0.5029486009565798 0.49220838989635535 0.514119726033024 0.02227713427713428 0.01772549335049335 0.027349259974259972 0.05884513084513086 0.04750401115401116 0.07049715787215788 18.302241956241957 17.049701190476192 19.5555055984556 0.020694122694122692 0.01873700128700129 0.022657818532818536 0.06109223509223509 0.05726682754182755 0.06464848777348778 16.701135564135562 15.928820785070785 17.462392299442303
6 M3_sourceTime text_audio 100 1.0 0.13965087270742027 0.11086805911581128 0.17086291475965698 37 0.07104170450923547 0.04617817302421454 0.09790063937091513 0.06860916809770604 0.05853303224689769 0.07936001767631712 0.6837595762555869 0.6552860960409499 0.712458437802354 0.07891848991848992 0.06314130201630203 0.09810974903474903 0.17487301587301587 0.14614791934791935 0.2047123552123552 11.77520634920635 10.68862012012012 12.995895302445302 0.055659373659373665 0.04583983268983269 0.06627339124839127 0.13317975117975117 0.11395983483483482 0.15161607679107675 12.579315744315746 11.378763942513944 13.799107142857137
7 M3_sourceTime text_vision 100 1.0 0.018218856013835664 -0.0019383159056188255 0.03824574691640572 37 0.01843536877672055 -0.0031358283705724124 0.0398007043003149 -0.00021651274608931568 -0.0033959447400838593 0.0026867472750431616 0.5073603928776801 0.5013866661304163 0.5138052809268234 0.02189274989274989 0.017497951522951527 0.026569176319176314 0.06272629772629773 0.053270045045045046 0.07303394465894468 18.673179751179752 17.663443983268984 19.694041184041183 0.01792406692406693 0.015639414414414417 0.02020035392535392 0.052900900900900896 0.04713513513513515 0.058622297297297295 17.553818961818966 16.37964121621622 18.782360049335054
8 M4_sourceTime audio_vision 100 1.0 0.00960363132716526 -0.026355777083579862 0.048860038747030224 37 0.008486620442388824 -0.028060019295580236 0.04573340321810003 0.0011170108847764362 -0.0023584936828478765 0.004587398100880476 0.5134115260777495 0.4984700587080242 0.529333335144605 0.02358944658944659 0.01981911196911197 0.027432882882882884 0.06169626769626772 0.05163449163449164 0.07249851994851997 17.65543586443586 16.274862580437578 19.030651254826253 0.0226001716001716 0.019407979407979413 0.026030244530244533 0.06180137280137281 0.05271195838695839 0.07151898326898326 17.76484513084513 16.36237923637924 19.1801465036465
9 M4_sourceTime text_audio 100 1.0 0.06588486519170021 0.03945950347046453 0.09289460044348818 37 0.042297942433741985 0.019241170794713554 0.06466475258021885 0.02358692279788703 0.01496262407228283 0.032121939048117505 0.6160321730354975 0.5843945260158524 0.6481647989887683 0.037331617331617335 0.03060465465465466 0.043963170313170305 0.10766709566709566 0.09256888674388675 0.12209738309738306 13.641798798798801 12.305344755469758 15.062401372801373 0.03636851136851138 0.027877606177606182 0.04520361432861432 0.10497468897468898 0.08676978764478764 0.12320139425139427 13.546082368082368 12.237779311454313 14.863573423423423
10 M4_sourceTime text_vision 100 1.0 0.002750558023057728 -0.02300323787854146 0.028440979736176288 37 -0.0017949415156420013 -0.03015163323847945 0.026173505636594735 0.004545499502936101 0.00029588325187894103 0.009067848319586198 0.5036754941681804 0.4941306089880723 0.5134271535022865 0.024956670956670958 0.020241580866580868 0.030214564564564565 0.06877262977262978 0.056675429000429 0.08342528957528957 17.827770913770916 16.393623906048905 19.236555115830114 0.018706134706134706 0.01641861861861862 0.020702370227370228 0.0541947661947662 0.049403925353925356 0.05915512655512656 18.49487387387387 17.468441087516084 19.530955534105537
11 TSFA-global audio_vision 100 1.0 0.058836026214662884 0.009433591336071618 0.10898509821607989 37 0.0299474197815315 -0.019026566930953497 0.07779842395164258 0.02888860643662733 0.019019952453169467 0.039115650392871 0.5811011281037877 0.5633139432698278 0.5997097862395071 0.05360531960531961 0.04721345988845988 0.06024396181896182 0.1473792363792364 0.1310375482625483 0.16456441441441436 12.399526383526384 11.223109202059202 13.630522844272843 0.026169884169884174 0.023153153153153153 0.02972972972972973 0.072981552981553 0.06513077005577005 0.0819059738309738 16.35457228657229 15.289285274560278 17.390430094380097
12 TSFA-global text_audio 100 1.0 0.6151585970459093 0.592303545010855 0.6363802687550169 37 0.3307069785888169 0.30723476551075196 0.3557131022926266 0.28445161780264955 0.2568491350423748 0.313431063751506 0.888037095225925 0.8713038621448196 0.9025196944665029 0.1913303303303303 0.17120793650793648 0.21155896610896607 0.44082539682539684 0.4059675246675246 0.4759347168597168 4.0553886743886745 3.421013191763192 4.85362648005148 0.13350965250965252 0.11927451737451739 0.1480651222651223 0.3463878163878164 0.31618583226083224 0.37751728871728873 6.496407550407551 5.68826003861004 7.341266816816816
13 TSFA-global text_vision 100 1.0 0.05193602225629566 0.00557772744820211 0.09913521469205716 37 0.018522642509243077 -0.031060642072022387 0.06214130753726415 0.03341337973851245 0.023362873667806743 0.044610356791469324 0.5860619295207061 0.5682458135804879 0.6058382392698549 0.0624924924924925 0.05556150793650794 0.0695357679107679 0.1712166452166452 0.15426744959244962 0.18985087945087942 10.319833118833117 9.234121578721577 11.574682368082371 0.027978978978978986 0.02543226083226084 0.030679751179751174 0.08042342342342342 0.07197972972972973 0.08928452380952381 16.16939510939511 14.841324195624193 17.51072467824968
14 TSFA-main audio_vision 100 1.0 0.014167320872974634 -0.010923848048597458 0.039202572867964074 37 0.00572396734721069 -0.022427993542694716 0.03323088554438458 0.008443353525896292 0.0026837436866675632 0.015297084395711374 0.5168712160108436 0.5063480574752184 0.5274386673004691 0.02177691977691978 0.01702814242814243 0.026475278850278845 0.059097383097383105 0.0487433075933076 0.06956501501501502 17.385808665808668 16.533373498498495 18.1887395002145 0.020783783783783785 0.017991763191763196 0.02361460746460746 0.059234234234234226 0.0514413770913771 0.06677416344916343 17.36473530673531 16.14145280995281 18.719550675675674
15 TSFA-main text_audio 100 1.0 0.44300997549055066 0.41801447102011324 0.4684682627939124 37 0.23093871736075175 0.2083505803346634 0.254020971881992 0.21207125773569485 0.18952878732104791 0.23568765088817542 0.8444317785680818 0.8258805004947956 0.8617723152745259 0.12499742599742601 0.11379501287001287 0.13663513513513512 0.3075298155298155 0.2829487880737881 0.3312437258687259 6.916232089232089 6.001308118833121 7.864674903474903 0.1002827112827113 0.08834907764907765 0.11306663449163447 0.2701518661518662 0.24654092664092664 0.29654121621621626 7.918441441441442 6.990601801801802 8.871822072072073
16 TSFA-main text_vision 100 1.0 0.0174827435109615 -0.014795200237153264 0.0498013264192671 37 0.007209132154900075 -0.027171939497727737 0.040232186743695476 0.010273611322660614 0.0044551992623352394 0.01675656997344102 0.5184451704744258 0.5070176356542713 0.5300536991588852 0.02143886743886744 0.017653732303732305 0.02514813384813385 0.06253453453453454 0.05233642213642214 0.07386306306306306 17.168172458172457 16.094917084942086 18.22205849420849 0.021965250965250968 0.019243018018018018 0.024749935649935648 0.06071342771342771 0.05453082368082369 0.0671651866151866 16.860575718575717 15.960335660660665 17.848634663234666
17 TSFA-multiply audio_vision 100 1.0 0.011799700871433188 -0.013416462369410848 0.03714225595848734 37 0.003975713585776283 -0.023566553326942314 0.029399206363613498 0.007823987230021269 0.002012223205404298 0.014442908524818563 0.5172232899693006 0.5074870466258432 0.5276387511371553 0.020090519090519093 0.01639286786786787 0.023955534105534106 0.05755426855426856 0.04768575718575719 0.06837712355212355 17.54742471042471 16.713855855855854 18.3979314993565 0.020211497211497214 0.017404579579579582 0.022875203775203776 0.058004719004719 0.05067953667953669 0.06470415057915058 16.951271986271987 15.84848443800944 18.024721492921493
18 TSFA-multiply text_audio 100 1.0 0.43704115381661734 0.4115881279842063 0.46182952157855794 37 0.22681150353265547 0.20526582196287743 0.2506810446160017 0.21022965143079514 0.186519824604155 0.23452360996737565 0.8422674450666473 0.8232177948179944 0.8604417612806575 0.12100386100386103 0.10705222007722007 0.13455019305019308 0.3025752895752896 0.27550450450450453 0.3296434684684685 7.047516087516087 6.080864478764479 8.009039800514799 0.10672972972972973 0.09416164736164735 0.11945048262548263 0.2784924924924925 0.2542594487344487 0.3048471042471042 7.982056199056198 7.081995849420849 8.870854268554268
19 TSFA-multiply text_vision 100 1.0 0.019027262118925484 -0.011798013240855429 0.05015302798114453 37 0.008701149625835541 -0.025909994276746277 0.04330936423482336 0.010326112528868553 0.004384441093969887 0.016339780587871982 0.5203762987911925 0.5091359831811294 0.5319951762695445 0.021473616473616473 0.017847393822393823 0.025207303732303732 0.06058687258687259 0.05200009652509653 0.06891801801801802 17.13536293436293 16.116098712998713 18.12695836550836 0.022200772200772205 0.01949655727155727 0.02511743886743887 0.05943286143286144 0.053154193479193486 0.0655993993993994 16.844067353067352 15.867397082797082 17.840103292578295
20 TSFA-random audio_vision 100 20.0 0.0003412659092388615 -0.012525598498225196 0.01250697673623247 37 -0.002757302584965167 -0.017743520338419794 0.011022231272467209 0.0030985684875587003 0.0007254310763673855 0.005789096312349675 0.520877531278429 0.5157918074214983 0.5265877221121911 0.03191679536679536 0.0298980518018018 0.0340041924066924 0.09458826683826685 0.08867857142857143 0.10115269626769628 15.031224281424278 14.405575359287857 15.620949209030458 0.021313685113685112 0.02049507185757186 0.022244750643500646 0.06297726297726298 0.06123326576576577 0.0648488315100815 16.687722222222224 15.855059311990566 17.524219135027884
21 TSFA-random text_audio 100 20.0 0.5176953808174132 0.49329636456565923 0.5421801408262926 37 0.27252678952364234 0.24728572903165366 0.2993041891233309 0.24516859096668722 0.22255121880186324 0.26788503176686185 0.8510175559155282 0.8342504583359005 0.8671632742516454 0.13294581724581728 0.12157137548262546 0.14623681306306305 0.35170619905619904 0.32670355051480054 0.37593875482625483 5.308989296439297 4.696646238202488 5.932726235521235 0.11373277563277563 0.1038839065851566 0.12520446160446164 0.2932850278850279 0.2702927622265122 0.3166170704633205 7.3795730802230795 6.716822666237666 8.059022419026167
22 TSFA-random text_vision 100 20.0 0.01558142443491039 -0.004607028207164052 0.0340684453611749 37 0.01131799393027936 -0.01173588358127808 0.03077004442264562 0.0042634304992695575 0.0006321330527004355 0.008231108258869975 0.523311412590402 0.5182214701491481 0.5286413477330866 0.03621218361218361 0.03320817889317888 0.03949472597597597 0.1053566495066495 0.09686440047190047 0.11374875321750319 14.090841570141572 13.219035881595882 14.95690251287001 0.021406113256113257 0.02049555823680824 0.022317504826254827 0.06158863148863149 0.05928467771342772 0.06384178035178034 16.919727498927497 15.876626134706132 18.00174117599742
Binary file not shown.

After

Width:  |  Height:  |  Size: 71 KiB

@@ -0,0 +1,7 @@
method,control,pair,clip_count,video_id_count,matched_vs_shifted_auc_video_macro_mean,matched_vs_shifted_auc_ci95_low,matched_vs_shifted_auc_ci95_high,same_minus_shifted_margin_video_macro_mean,same_minus_shifted_margin_ci95_low,same_minus_shifted_margin_ci95_high,exact_r1_left_to_right_video_macro_mean,exact_r1_left_to_right_ci95_low,exact_r1_left_to_right_ci95_high,within_pm1_r1_left_to_right_video_macro_mean,within_pm1_r1_left_to_right_ci95_low,within_pm1_r1_left_to_right_ci95_high
M4_sourceTime,independent_within_clip_permutation,audio_vision,100,37,0.49986099215667035,0.49931733210509144,0.5003794440377403,-0.0003183416417442903,-0.0006530043625340906,-3.3575496045077684e-05,0.02070763620763621,0.019744644465894464,0.021695343736593736,0.059731810381810385,0.05825279922779924,0.06113831724581724
M4_sourceTime,independent_within_clip_permutation,text_audio,100,37,0.4992583332049756,0.4971305291708881,0.5012287335156442,-0.0004793830947098817,-0.0017547874540129774,0.0007121573235354092,0.01981724581724582,0.018432171814671817,0.021092101029601035,0.05907595452595453,0.057171570141570136,0.06100200343200343
M4_sourceTime,independent_within_clip_permutation,text_vision,100,37,0.4995061476417528,0.49898081806837713,0.5000416571403756,-0.0001465278630284633,-0.00048526501402030057,0.00016435085172832273,0.01881887601887602,0.01782144305019305,0.01977210102960103,0.058243114543114545,0.05672013513513514,0.059936018339768325
TSFA-main,independent_within_clip_permutation,audio_vision,100,37,0.5002577534839172,0.4994096693338,0.5011686280803744,0.00012270166467700342,-0.0003552724864745554,0.0005955004762808745,0.019852552552552555,0.0187075,0.020917128378378377,0.059261347061347065,0.05789174603174604,0.060541009223509225
TSFA-main,independent_within_clip_permutation,text_audio,100,37,0.5004550297305284,0.4981946165487771,0.5026825985666993,0.0006103920254983002,-0.0024510630057079466,0.0037619848284550215,0.01998494208494209,0.019009004182754183,0.02102998873873874,0.05908569283569285,0.05776156960531959,0.06049437419562419
TSFA-main,independent_within_clip_permutation,text_vision,100,37,0.5000126508394793,0.49939359917466397,0.5006472318623715,-0.00016505359922095624,-0.0005699285037343316,0.00024382621948318463,0.019825032175032176,0.01893382561132561,0.020605167310167313,0.059049335049335065,0.05775769841269841,0.06029238685113685
1 method control pair clip_count video_id_count matched_vs_shifted_auc_video_macro_mean matched_vs_shifted_auc_ci95_low matched_vs_shifted_auc_ci95_high same_minus_shifted_margin_video_macro_mean same_minus_shifted_margin_ci95_low same_minus_shifted_margin_ci95_high exact_r1_left_to_right_video_macro_mean exact_r1_left_to_right_ci95_low exact_r1_left_to_right_ci95_high within_pm1_r1_left_to_right_video_macro_mean within_pm1_r1_left_to_right_ci95_low within_pm1_r1_left_to_right_ci95_high
2 M4_sourceTime independent_within_clip_permutation audio_vision 100 37 0.49986099215667035 0.49931733210509144 0.5003794440377403 -0.0003183416417442903 -0.0006530043625340906 -3.3575496045077684e-05 0.02070763620763621 0.019744644465894464 0.021695343736593736 0.059731810381810385 0.05825279922779924 0.06113831724581724
3 M4_sourceTime independent_within_clip_permutation text_audio 100 37 0.4992583332049756 0.4971305291708881 0.5012287335156442 -0.0004793830947098817 -0.0017547874540129774 0.0007121573235354092 0.01981724581724582 0.018432171814671817 0.021092101029601035 0.05907595452595453 0.057171570141570136 0.06100200343200343
4 M4_sourceTime independent_within_clip_permutation text_vision 100 37 0.4995061476417528 0.49898081806837713 0.5000416571403756 -0.0001465278630284633 -0.00048526501402030057 0.00016435085172832273 0.01881887601887602 0.01782144305019305 0.01977210102960103 0.058243114543114545 0.05672013513513514 0.059936018339768325
5 TSFA-main independent_within_clip_permutation audio_vision 100 37 0.5002577534839172 0.4994096693338 0.5011686280803744 0.00012270166467700342 -0.0003552724864745554 0.0005955004762808745 0.019852552552552555 0.0187075 0.020917128378378377 0.059261347061347065 0.05789174603174604 0.060541009223509225
6 TSFA-main independent_within_clip_permutation text_audio 100 37 0.5004550297305284 0.4981946165487771 0.5026825985666993 0.0006103920254983002 -0.0024510630057079466 0.0037619848284550215 0.01998494208494209 0.019009004182754183 0.02102998873873874 0.05908569283569285 0.05776156960531959 0.06049437419562419
7 TSFA-main independent_within_clip_permutation text_vision 100 37 0.5000126508394793 0.49939359917466397 0.5006472318623715 -0.00016505359922095624 -0.0005699285037343316 0.00024382621948318463 0.019825032175032176 0.01893382561132561 0.020605167310167313 0.059049335049335065 0.05775769841269841 0.06029238685113685
@@ -0,0 +1,26 @@
left_method,right_method,metric_family,pair,metric,clip_count,video_id_count,left_minus_right_video_macro_mean,ci95_low,ci95_high
TSFA-main,M4_sourceTime,overall,all_pairs,content_auc_mean,100,37,0.08220965725730794,0.07113723316940294,0.09569307390121884
TSFA-main,M4_sourceTime,overall,all_pairs,canonical_pairwise_time_mae_mean,100,37,0.005933525736358895,0.005308267109580502,0.006529948953558749
TSFA-main,M4_sourceTime,content_pair,text_audio,matched_vs_shifted_auc,100,37,0.22839960553258426,0.19753921496296828,0.2586728798778832
TSFA-main,M4_sourceTime,content_pair,text_vision,matched_vs_shifted_auc,100,37,0.014769676306245455,0.00615762201696377,0.024147940037301717
TSFA-main,M4_sourceTime,content_pair,audio_vision,matched_vs_shifted_auc,100,37,0.003459689933094183,-0.0127482317281187,0.01864119744744742
TSFA-main,TSFA-random,overall,all_pairs,content_auc_mean,100,37,-0.00515277824366921,-0.011173873250387557,0.0007255270180419965
TSFA-main,TSFA-random,overall,all_pairs,canonical_pairwise_time_mae_mean,100,37,-0.18037076895618454,-0.18509186583110665,-0.17551767054171094
TSFA-main,TSFA-random,content_pair,text_audio,matched_vs_shifted_auc,100,37,-0.006585777347446219,-0.015532873664531726,0.0009897398589461357
TSFA-main,TSFA-random,content_pair,text_vision,matched_vs_shifted_auc,100,37,-0.00486624211597616,-0.014786038407090595,0.004877792533937811
TSFA-main,TSFA-random,content_pair,audio_vision,matched_vs_shifted_auc,100,37,-0.004006315267585209,-0.013151033853001922,0.005371947005658802
TSFA-main,TSFA-global,overall,all_pairs,content_auc_mean,100,37,-0.05848399593235586,-0.06921582521604686,-0.04751669353941074
TSFA-main,TSFA-global,overall,all_pairs,canonical_pairwise_time_mae_mean,100,37,-0.17289980508939898,-0.17836035039600534,-0.16750727197744386
TSFA-main,TSFA-global,content_pair,text_audio,matched_vs_shifted_auc,100,37,-0.043605316657843264,-0.053833684686776434,-0.03223812133752294
TSFA-main,TSFA-global,content_pair,text_vision,matched_vs_shifted_auc,100,37,-0.06761675904628031,-0.0849899430490521,-0.050819432782896884
TSFA-main,TSFA-global,content_pair,audio_vision,matched_vs_shifted_auc,100,37,-0.06422991209294399,-0.0784464424435834,-0.050670400781670726
TSFA-multiply,TSFA-main,overall,all_pairs,content_auc_mean,100,37,3.9622924596328686e-05,-0.0011158515658515594,0.0011833296608728741
TSFA-multiply,TSFA-main,overall,all_pairs,canonical_pairwise_time_mae_mean,100,37,-0.0025367852317495707,-0.0027293981115725143,-0.0023532245292300245
TSFA-multiply,TSFA-main,content_pair,text_audio,matched_vs_shifted_auc,100,37,-0.0021643335014345675,-0.003420222806393028,-0.0009433171987527146
TSFA-multiply,TSFA-main,content_pair,text_vision,matched_vs_shifted_auc,100,37,0.0019311283167666208,-3.923633930282419e-05,0.003844595478837497
TSFA-multiply,TSFA-main,content_pair,audio_vision,matched_vs_shifted_auc,100,37,0.00035207395845693497,-0.001707745444912998,0.0025183332791377994
TSFA-main,M3_noSourceTime,overall,all_pairs,content_auc_mean,100,37,-0.08134204496416375,-0.09018397700115897,-0.07252863456636728
TSFA-main,M3_noSourceTime,overall,all_pairs,canonical_pairwise_time_mae_mean,100,37,-0.18092685065902986,-0.18650572139607072,-0.17507314724507567
TSFA-main,M3_noSourceTime,content_pair,text_audio,matched_vs_shifted_auc,100,37,-0.11151433985277072,-0.126396370047384,-0.09652581659832322
TSFA-main,M3_noSourceTime,content_pair,text_vision,matched_vs_shifted_auc,100,37,-0.08788246281996281,-0.10456414269512673,-0.07031106495933623
TSFA-main,M3_noSourceTime,content_pair,audio_vision,matched_vs_shifted_auc,100,37,-0.04462933221975775,-0.054448958632834964,-0.0345335877670984
1 left_method right_method metric_family pair metric clip_count video_id_count left_minus_right_video_macro_mean ci95_low ci95_high
2 TSFA-main M4_sourceTime overall all_pairs content_auc_mean 100 37 0.08220965725730794 0.07113723316940294 0.09569307390121884
3 TSFA-main M4_sourceTime overall all_pairs canonical_pairwise_time_mae_mean 100 37 0.005933525736358895 0.005308267109580502 0.006529948953558749
4 TSFA-main M4_sourceTime content_pair text_audio matched_vs_shifted_auc 100 37 0.22839960553258426 0.19753921496296828 0.2586728798778832
5 TSFA-main M4_sourceTime content_pair text_vision matched_vs_shifted_auc 100 37 0.014769676306245455 0.00615762201696377 0.024147940037301717
6 TSFA-main M4_sourceTime content_pair audio_vision matched_vs_shifted_auc 100 37 0.003459689933094183 -0.0127482317281187 0.01864119744744742
7 TSFA-main TSFA-random overall all_pairs content_auc_mean 100 37 -0.00515277824366921 -0.011173873250387557 0.0007255270180419965
8 TSFA-main TSFA-random overall all_pairs canonical_pairwise_time_mae_mean 100 37 -0.18037076895618454 -0.18509186583110665 -0.17551767054171094
9 TSFA-main TSFA-random content_pair text_audio matched_vs_shifted_auc 100 37 -0.006585777347446219 -0.015532873664531726 0.0009897398589461357
10 TSFA-main TSFA-random content_pair text_vision matched_vs_shifted_auc 100 37 -0.00486624211597616 -0.014786038407090595 0.004877792533937811
11 TSFA-main TSFA-random content_pair audio_vision matched_vs_shifted_auc 100 37 -0.004006315267585209 -0.013151033853001922 0.005371947005658802
12 TSFA-main TSFA-global overall all_pairs content_auc_mean 100 37 -0.05848399593235586 -0.06921582521604686 -0.04751669353941074
13 TSFA-main TSFA-global overall all_pairs canonical_pairwise_time_mae_mean 100 37 -0.17289980508939898 -0.17836035039600534 -0.16750727197744386
14 TSFA-main TSFA-global content_pair text_audio matched_vs_shifted_auc 100 37 -0.043605316657843264 -0.053833684686776434 -0.03223812133752294
15 TSFA-main TSFA-global content_pair text_vision matched_vs_shifted_auc 100 37 -0.06761675904628031 -0.0849899430490521 -0.050819432782896884
16 TSFA-main TSFA-global content_pair audio_vision matched_vs_shifted_auc 100 37 -0.06422991209294399 -0.0784464424435834 -0.050670400781670726
17 TSFA-multiply TSFA-main overall all_pairs content_auc_mean 100 37 3.9622924596328686e-05 -0.0011158515658515594 0.0011833296608728741
18 TSFA-multiply TSFA-main overall all_pairs canonical_pairwise_time_mae_mean 100 37 -0.0025367852317495707 -0.0027293981115725143 -0.0023532245292300245
19 TSFA-multiply TSFA-main content_pair text_audio matched_vs_shifted_auc 100 37 -0.0021643335014345675 -0.003420222806393028 -0.0009433171987527146
20 TSFA-multiply TSFA-main content_pair text_vision matched_vs_shifted_auc 100 37 0.0019311283167666208 -3.923633930282419e-05 0.003844595478837497
21 TSFA-multiply TSFA-main content_pair audio_vision matched_vs_shifted_auc 100 37 0.00035207395845693497 -0.001707745444912998 0.0025183332791377994
22 TSFA-main M3_noSourceTime overall all_pairs content_auc_mean 100 37 -0.08134204496416375 -0.09018397700115897 -0.07252863456636728
23 TSFA-main M3_noSourceTime overall all_pairs canonical_pairwise_time_mae_mean 100 37 -0.18092685065902986 -0.18650572139607072 -0.17507314724507567
24 TSFA-main M3_noSourceTime content_pair text_audio matched_vs_shifted_auc 100 37 -0.11151433985277072 -0.126396370047384 -0.09652581659832322
25 TSFA-main M3_noSourceTime content_pair text_vision matched_vs_shifted_auc 100 37 -0.08788246281996281 -0.10456414269512673 -0.07031106495933623
26 TSFA-main M3_noSourceTime content_pair audio_vision matched_vs_shifted_auc 100 37 -0.04462933221975775 -0.054448958632834964 -0.0345335877670984
@@ -0,0 +1,260 @@
{
"created_utc": "2026-09-23T14:26:26.904353+00:00",
"experiment": "TSFA: Temporal-Semantic Factorized Alignment",
"sample_count": 100,
"fold_count": 5,
"heldout_partition_count": 100,
"unique_video_id_count": 37,
"folds": [
{
"fold": 1,
"train_count": 80,
"heldout_count": 20,
"train_video_id_count": 31,
"heldout_video_id_count": 6,
"video_id_overlap": [],
"validation_sample_ids": [
"-3g5yACwYnA/13",
"-3g5yACwYnA/3",
"-3g5yACwYnA/2",
"-3g5yACwYnA/9",
"-egA8-b7-3M/26",
"-egA8-b7-3M/17",
"-egA8-b7-3M/18",
"-egA8-b7-3M/16",
"-egA8-b7-3M/13",
"-egA8-b7-3M/1",
"-egA8-b7-3M/6",
"-egA8-b7-3M/9",
"-egA8-b7-3M/20",
"-I_e4mIh0yE/1",
"-I_e4mIh0yE/3",
"-qDkUB0GgYY/6",
"-9y-fZ3swSY/0",
"-9y-fZ3swSY/4",
"-9y-fZ3swSY/8",
"-UUCSKoHeMA/0"
]
},
{
"fold": 2,
"train_count": 80,
"heldout_count": 20,
"train_video_id_count": 30,
"heldout_video_id_count": 7,
"video_id_overlap": [],
"validation_sample_ids": [
"-THoVjtIkeU/12",
"-THoVjtIkeU/2",
"-THoVjtIkeU/6",
"-UuX1xuaiiE/1",
"-UuX1xuaiiE/0",
"-UuX1xuaiiE/3",
"-UuX1xuaiiE/6",
"-lzEya4AM_4/5",
"-lzEya4AM_4/6",
"-mqbVkbCndg/0",
"-wny0OAz3g8/1",
"-wny0OAz3g8/0",
"-wny0OAz3g8/3",
"-wny0OAz3g8/2",
"-wny0OAz3g8/5",
"-wny0OAz3g8/7",
"-wny0OAz3g8/9",
"-571d8cVauQ/0",
"-571d8cVauQ/5",
"-HeZS2-Prhc/2"
]
},
{
"fold": 3,
"train_count": 80,
"heldout_count": 20,
"train_video_id_count": 29,
"heldout_video_id_count": 8,
"video_id_overlap": [],
"validation_sample_ids": [
"-HwX2H8Z4hY/2",
"-HwX2H8Z4hY/5",
"-HwX2H8Z4hY/6",
"-HwX2H8Z4hY/9",
"-aqamKhZ1Ec/0",
"-vxjVxOeScU/4",
"-wMB_hJL-3o/7",
"-hnBHBN8p5A/7",
"-hnBHBN8p5A/6",
"-AUZQgSxyPQ/2",
"-RfYyzHpjk4/11",
"-RfYyzHpjk4/8",
"-RfYyzHpjk4/2",
"-s9qJ7ATP7w/1",
"-s9qJ7ATP7w/0",
"-s9qJ7ATP7w/5",
"-s9qJ7ATP7w/4",
"-s9qJ7ATP7w/7",
"-s9qJ7ATP7w/6",
"-s9qJ7ATP7w/8"
]
},
{
"fold": 4,
"train_count": 80,
"heldout_count": 20,
"train_video_id_count": 29,
"heldout_video_id_count": 8,
"video_id_overlap": [],
"validation_sample_ids": [
"-aNfi7CP8vM/7",
"-iRBcNs9oI8/3",
"-iRBcNs9oI8/7",
"-iRBcNs9oI8/6",
"-iRBcNs9oI8/9",
"-iRBcNs9oI8/8",
"-t217m2on-s/2",
"-t217m2on-s/7",
"-tPCytz4rww/11",
"-tPCytz4rww/10",
"-tPCytz4rww/12",
"-tPCytz4rww/16",
"-tPCytz4rww/18",
"-UacrmKiTn4/10",
"-UacrmKiTn4/4",
"-uywlfIYOS8/4",
"-6rXp3zJ3kc/8",
"-ri04Z7vwnc/0",
"-ri04Z7vwnc/2",
"-ri04Z7vwnc/5"
]
},
{
"fold": 5,
"train_count": 80,
"heldout_count": 20,
"train_video_id_count": 29,
"heldout_video_id_count": 8,
"video_id_overlap": [],
"validation_sample_ids": [
"-3nNcZdcdvU/5",
"-NFrJFQijFE/1",
"-NFrJFQijFE/2",
"-a55Q6RWvTA/3",
"-dxfTGcXJoc/1",
"-dxfTGcXJoc/0",
"-dxfTGcXJoc/2",
"-dxfTGcXJoc/6",
"-mJ2ud6oKI8/1",
"-mJ2ud6oKI8/2",
"-mJ2ud6oKI8/6",
"-mJ2ud6oKI8/9",
"-mJ2ud6oKI8/8",
"-tANM6ETl_M/3",
"-MeTTeMJBNc/0",
"-MeTTeMJBNc/13",
"-MeTTeMJBNc/7",
"-yRb-Jum7EQ/1",
"-yRb-Jum7EQ/5",
"-yRb-Jum7EQ/6"
]
}
],
"alignment_models_retrained": false,
"temporal_branch": "frozen M4_sourceTime D5 checkpoint per grouped fold",
"semantic_branch_trained": true,
"emotion_labels_used_for_alignment": false,
"feature_extractors_changed": false,
"feature_encoder_note": "Existing BERT text and DeiT image features are reused unchanged.",
"example_sample_id": "-3g5yACwYnA/13",
"parameters": {
"seed": 42,
"grid_size": 50,
"hidden_size": 128,
"heads_in_frozen_alignment": 4,
"delta_normalized_time": 0.1,
"hard_negative_offsets_slots": [
-5,
-3,
-2,
2,
3,
5
],
"semantic_epochs": 40,
"semantic_learning_rate": 0.001,
"local_contrastive_temperature": 0.1,
"content_probe_epochs": 40,
"content_probe_temperature": 0.1,
"reconstruction_probe_epochs": 40,
"shuffle_repeats": 20,
"random_window_repeats": 20,
"window_ablation_status": "delta sweep and 80-percent attention-mass window deferred until the seven-method MVP is reviewed"
},
"methods": [
"M3_noSourceTime",
"M3_sourceTime",
"M4_sourceTime",
"TSFA-main",
"TSFA-multiply",
"TSFA-random",
"TSFA-global"
],
"protocol": {
"semantic_query_key_inputs": "Frozen M4 attention-pooled text value output is the semantic query input; M4 pre-attention projected Audio/Vision content is the semantic key/value input. No explicit source-time code, absolute PE, slot ID, or latent positional vector is passed to semantic Q/K. Temporal selection can still encode time indirectly.",
"baseline_content_values": "M4 and M3 Audio/Vision use each checkpoint's actual MHA attention output. M3 Text uses attention-pooled pre-attention text features to remove its source-time query residual.",
"temporal_role": "Frozen M4 attention supplies candidate masks; TSFA-multiply additionally multiplies semantic probabilities by M4 temporal attention",
"local_contrastive": "same-slot positives with within-clip slot-offset negatives at +/-2, +/-3, +/-5; symmetric Text-Audio and Text-Vision loss",
"random_candidate": "same delta-width windows with independently sampled centers at inference; 20 held-out random draws, one train draw for the train-only probe",
"global_candidate": "all valid source positions available to semantic attention",
"content_probe": "same train-only symmetric within-clip InfoNCE probe applied to each method; held-out negatives are positions more than two slots away",
"content_shuffle": "independently permute each modality's 50 projected content rows within each held-out clip; M4 and TSFA-main",
"reconstruction": "fold-train decoder predicts one aligned content representation from the other two; shift one partner by signed offsets +/-1,2,5,10 on identical target support",
"confidence_intervals": "95-percent video_id-cluster bootstrap; 2,000 repetitions for CSV summaries"
},
"input_paths": {
"feature_dir": "/home/gloamxun/modeling_zhaocui/deep_learning/Q1/outputs/q1_features/features",
"feature_manifest": "/home/gloamxun/modeling_zhaocui/deep_learning/Q1/outputs/audit/manifest.csv",
"grouped_splits": "/home/gloamxun/modeling_zhaocui/deep_learning/Q1/outputs/method_comparison/splits.json",
"checkpoint_root": "/home/gloamxun/modeling_zhaocui/deep_learning/Q1/outputs/alignment_debug/heldout",
"m3_m4_checkpoint_variants": [
"M3_noSourceTime",
"M3_sourceTime",
"M4_sourceTime"
]
},
"device": "cuda",
"gpu_name": "NVIDIA GeForce RTX 5070 Ti",
"python": "3.14.7",
"torch": "2.14.0+cu130",
"elapsed_seconds": 129.8167963027954,
"interpretation_limits": [
"Same-slot labels are a temporal training convention, not independent semantic ground truth.",
"Frozen M4 temporal attention was trained with a timestamp-derived Gaussian prior; time may enter content indirectly through candidate selection.",
"AUC, retrieval, and shifted similarity measure within-clip content matchability, not human event alignment.",
"Reconstruction predicts M4/TSFA pooled content values, not raw waveform or image pixels.",
"Five folds use one alignment-model seed; bootstrap intervals quantify video-group sampling uncertainty, not seed uncertainty.",
"Human event IoU/center error remains pending event annotation."
],
"input_sha256": {
"feature_manifest": "bc07d5fb1ff104144cf845abf856c4e04da3caadda0c05d4b798a64a581bb35a",
"grouped_splits": "ccdcab416488fd3d94b762696f70411ebcd02346417cb7f1a5be70e7a1a6cf4d",
"uv_lock": "28ae0c736eef130a380022da82c79c473f467e2cb8dad4c199a85fd0b1faadf1",
"features_combined": "ed99618a0f4ac943fd21c214a331aeff11ca6aa7419b48622ab25f8f8d8ae985",
"feature_file_count": 100,
"frozen_alignment_checkpoints": {
"fold_01/M3_noSourceTime": "e54088fbc4432f41e9a09b167cd90a752c6bf678461e4d4ca553e1a7eb23fadc",
"fold_01/M3_sourceTime": "377772c546ec1cb79359b4792de76f67df9700a1ab48f0dfc66451640733ef67",
"fold_01/M4_sourceTime": "ad2e5fb0db3a8891a831ad3d2c85dde59f0563bc41dc05fff10655589d37d1fc",
"fold_02/M3_noSourceTime": "2bd24ca12a8c6eb4335b54db5de7b0f9963752279f0bf4daa9c7641b251e00ae",
"fold_02/M3_sourceTime": "64eca35e24962055c8e2d1520f62af291f395f8243438bc1609925d62794c673",
"fold_02/M4_sourceTime": "f77c6a46ec6424b7a52907bca81a0842ec0066ba9dec38b8e10bfbbf455a1a62",
"fold_03/M3_noSourceTime": "76dff4f2643a4fd46dc6d7a5d7d6a74ce8e119da53fed81d98b4c22d90af0052",
"fold_03/M3_sourceTime": "f52e47bdbe8a3b3b360ad053a202570170f45cdfc6d07066f6e8558d31182125",
"fold_03/M4_sourceTime": "de6fcfbaa57eab745b6ccfd7a19860589fccaa30b497813a3afdecb15956ac6c",
"fold_04/M3_noSourceTime": "735354a326ab23a832370449862493bcab2452551858a6825c099c27ff982b58",
"fold_04/M3_sourceTime": "304c7287f4d777b27a5354e4316802cdc6de75dd1dd20a419314255b67618a4e",
"fold_04/M4_sourceTime": "d0e50814377b5b98fbf70c3ca48431756b22391823629dab74c3192b8bfe1a3d",
"fold_05/M3_noSourceTime": "0f8d57bae2af04e0a7432a42a68b02b09c60bac4da68d4b7595905a31e1c8fe2",
"fold_05/M3_sourceTime": "47bccf271f6046e695fd7ad7e16d9a018d944df5773b5a8e15af9be80bd9af41",
"fold_05/M4_sourceTime": "787e28df330b9891f8ff24c6522d808bd0a1bac269f29c6f11dff55b8ba95d4e"
}
}
}
Binary file not shown.

After

Width:  |  Height:  |  Size: 148 KiB

@@ -0,0 +1,25 @@
method,target_modality,shifted_modality,abs_delta,clip_count,video_id_count,aligned_mae_same_support_video_macro_mean,aligned_mae_same_support_ci95_low,aligned_mae_same_support_ci95_high,shifted_mae_video_macro_mean,shifted_mae_ci95_low,shifted_mae_ci95_high,gain_shift_minus_aligned_video_macro_mean,gain_shift_minus_aligned_ci95_low,gain_shift_minus_aligned_ci95_high
M4_sourceTime,audio,vision,1,100,37,0.10648079388657068,0.09498403317711335,0.11971105591350258,0.10648057482407897,0.09536464065503321,0.1203203971314147,-2.190624917080248e-07,-3.002898652523641e-06,2.836719715064019e-06
M4_sourceTime,audio,vision,10,100,37,0.10608212038697232,0.09495647062817791,0.11900391438347258,0.10613144132552779,0.09468061707438626,0.12023388864214406,4.932093855548141e-05,-0.000138402270899901,0.0002644478668511099
M4_sourceTime,audio,vision,2,100,37,0.10634162971084272,0.0951935955060543,0.12024532359561485,0.10634258744158634,0.09515405380473113,0.12009526039769572,9.577307436201306e-07,-1.0696270178279515e-05,1.3246172074194955e-05
M4_sourceTime,audio,vision,5,100,37,0.1060977863330044,0.09541860889488209,0.11980911076375722,0.1061057905801574,0.09548577428451295,0.1195104905764324,8.004247152986875e-06,-5.808767217863869e-05,7.687466521118136e-05
M4_sourceTime,text,vision,1,100,37,0.3524187293592465,0.3221663890648929,0.3832589358215371,0.35242382845190495,0.3240157492065436,0.38444638677967397,5.09909265852445e-06,-8.073796445031578e-06,2.1147349556082562e-05
M4_sourceTime,text,vision,10,100,37,0.34596756801060324,0.31742921319489337,0.37868989748536663,0.34664258137473164,0.31755937971955817,0.3774341924067551,0.0006750133641284713,-0.0002579232627475584,0.0017880953921477383
M4_sourceTime,text,vision,2,100,37,0.35132473441234785,0.32286859659712197,0.38345827370335456,0.3513478712829614,0.3224631958370705,0.38120628674285656,2.3136870613531815e-05,-2.753329515642539e-05,8.594694638849325e-05
M4_sourceTime,text,vision,5,100,37,0.3485910437432214,0.3205305428072932,0.3803397981259862,0.348778251139507,0.3203935584532843,0.3815424183815085,0.00018720739628561882,-9.496389538631425e-05,0.0005151327779809838
M4_sourceTime,vision,audio,1,100,37,0.2231766281827664,0.20607600232587064,0.24236390166586727,0.22317987500195693,0.20554712626708696,0.2422070137882051,3.246819190537148e-06,-2.2809803989287967e-06,9.420666630299547e-06
M4_sourceTime,vision,audio,10,100,37,0.22307425242673293,0.20607910972755006,0.24121378121482856,0.22325032634755695,0.2067523672321235,0.24305877808947715,0.00017607392082394369,-0.00022312340428489366,0.0006277892943623485
M4_sourceTime,vision,audio,2,100,37,0.22316048174652228,0.20552624977176193,0.24224169822075334,0.22317257495856702,0.2061885128990303,0.2429836954945456,1.209321204478625e-05,-9.359554453552767e-06,3.6892148305301144e-05
M4_sourceTime,vision,audio,5,100,37,0.2231394299643585,0.20618209030565493,0.24110680100715395,0.22321575703646734,0.20586565202003376,0.2427468816852832,7.632707210888122e-05,-4.880731381197847e-05,0.0002121602112554407
TSFA-main,audio,vision,1,100,37,0.24988408831952214,0.22469871818983359,0.27840707807349385,0.24988695082411724,0.22501213166380882,0.2790530408493829,2.8625045951371134e-06,-1.742456965739765e-05,2.272696183944069e-05
TSFA-main,audio,vision,10,100,37,0.24386757700414372,0.2190698984802626,0.2733444834273454,0.2444317758278318,0.21902110992468563,0.27219107996658004,0.000564198823688062,-0.00022641418153737327,0.001395855642402091
TSFA-main,audio,vision,2,100,37,0.24895100756913924,0.22416956525924045,0.27810367903911887,0.24896510513695155,0.22274126205659267,0.27930326741638245,1.4097567812305108e-05,-4.732088382019858e-05,7.63355802100572e-05
TSFA-main,audio,vision,5,100,37,0.2463452433058948,0.22097869036887005,0.2751587883843783,0.246497264087475,0.2214858425035993,0.2751850658218311,0.0001520207815802389,-0.00014739758679055951,0.0004554734628724275
TSFA-main,text,vision,1,100,37,0.3476741555274325,0.32148611245427566,0.3737497650390369,0.3476541439183967,0.32086054265937813,0.37359475283999777,-2.0011609035824965e-05,-6.729367620210594e-05,3.311202856174665e-05
TSFA-main,text,vision,10,100,37,0.33808821519346943,0.3123830481146549,0.36471129622609255,0.3382066619050689,0.31227041983330905,0.365546023769169,0.00011844671159944317,-0.001876128714584399,0.002146817830318111
TSFA-main,text,vision,2,100,37,0.3463484496062007,0.3194491144852671,0.372250999602138,0.34635784703948913,0.31847836298931187,0.3729820835754293,9.397433288411396e-06,-0.0001571169486153489,0.00020347758379682152
TSFA-main,text,vision,5,100,37,0.3427197299862511,0.31759720182711926,0.3687916805460675,0.34274824133142034,0.3161611675319563,0.37042456927433204,2.8511345169156534e-05,-0.0007572757773635113,0.0009358633216909969
TSFA-main,vision,audio,1,100,37,0.30385903824985366,0.2804930920345874,0.32447569376476193,0.3038967083492334,0.28237160721884863,0.32535285142726206,3.767009937983119e-05,2.2140309296097684e-05,5.2699988083822036e-05
TSFA-main,vision,audio,10,100,37,0.30352860244078117,0.2808613100727914,0.32450677675282535,0.30500827796376856,0.28273082537522215,0.32564273932073723,0.0014796755229874467,0.0004977929359841301,0.002395565682036881
TSFA-main,vision,audio,2,100,37,0.30382711134055096,0.2824193253098451,0.32531113133968326,0.3039569732806986,0.2827239277182888,0.3254230913166988,0.0001298619401475382,7.685713367700474e-05,0.00018430574637678292
TSFA-main,vision,audio,5,100,37,0.3037293283725082,0.2824902004786779,0.3239507917223273,0.30433064016570666,0.2829440499471156,0.32495928013412795,0.0006013117931984567,0.00028662674138598147,0.0009081425354786053
1 method target_modality shifted_modality abs_delta clip_count video_id_count aligned_mae_same_support_video_macro_mean aligned_mae_same_support_ci95_low aligned_mae_same_support_ci95_high shifted_mae_video_macro_mean shifted_mae_ci95_low shifted_mae_ci95_high gain_shift_minus_aligned_video_macro_mean gain_shift_minus_aligned_ci95_low gain_shift_minus_aligned_ci95_high
2 M4_sourceTime audio vision 1 100 37 0.10648079388657068 0.09498403317711335 0.11971105591350258 0.10648057482407897 0.09536464065503321 0.1203203971314147 -2.190624917080248e-07 -3.002898652523641e-06 2.836719715064019e-06
3 M4_sourceTime audio vision 10 100 37 0.10608212038697232 0.09495647062817791 0.11900391438347258 0.10613144132552779 0.09468061707438626 0.12023388864214406 4.932093855548141e-05 -0.000138402270899901 0.0002644478668511099
4 M4_sourceTime audio vision 2 100 37 0.10634162971084272 0.0951935955060543 0.12024532359561485 0.10634258744158634 0.09515405380473113 0.12009526039769572 9.577307436201306e-07 -1.0696270178279515e-05 1.3246172074194955e-05
5 M4_sourceTime audio vision 5 100 37 0.1060977863330044 0.09541860889488209 0.11980911076375722 0.1061057905801574 0.09548577428451295 0.1195104905764324 8.004247152986875e-06 -5.808767217863869e-05 7.687466521118136e-05
6 M4_sourceTime text vision 1 100 37 0.3524187293592465 0.3221663890648929 0.3832589358215371 0.35242382845190495 0.3240157492065436 0.38444638677967397 5.09909265852445e-06 -8.073796445031578e-06 2.1147349556082562e-05
7 M4_sourceTime text vision 10 100 37 0.34596756801060324 0.31742921319489337 0.37868989748536663 0.34664258137473164 0.31755937971955817 0.3774341924067551 0.0006750133641284713 -0.0002579232627475584 0.0017880953921477383
8 M4_sourceTime text vision 2 100 37 0.35132473441234785 0.32286859659712197 0.38345827370335456 0.3513478712829614 0.3224631958370705 0.38120628674285656 2.3136870613531815e-05 -2.753329515642539e-05 8.594694638849325e-05
9 M4_sourceTime text vision 5 100 37 0.3485910437432214 0.3205305428072932 0.3803397981259862 0.348778251139507 0.3203935584532843 0.3815424183815085 0.00018720739628561882 -9.496389538631425e-05 0.0005151327779809838
10 M4_sourceTime vision audio 1 100 37 0.2231766281827664 0.20607600232587064 0.24236390166586727 0.22317987500195693 0.20554712626708696 0.2422070137882051 3.246819190537148e-06 -2.2809803989287967e-06 9.420666630299547e-06
11 M4_sourceTime vision audio 10 100 37 0.22307425242673293 0.20607910972755006 0.24121378121482856 0.22325032634755695 0.2067523672321235 0.24305877808947715 0.00017607392082394369 -0.00022312340428489366 0.0006277892943623485
12 M4_sourceTime vision audio 2 100 37 0.22316048174652228 0.20552624977176193 0.24224169822075334 0.22317257495856702 0.2061885128990303 0.2429836954945456 1.209321204478625e-05 -9.359554453552767e-06 3.6892148305301144e-05
13 M4_sourceTime vision audio 5 100 37 0.2231394299643585 0.20618209030565493 0.24110680100715395 0.22321575703646734 0.20586565202003376 0.2427468816852832 7.632707210888122e-05 -4.880731381197847e-05 0.0002121602112554407
14 TSFA-main audio vision 1 100 37 0.24988408831952214 0.22469871818983359 0.27840707807349385 0.24988695082411724 0.22501213166380882 0.2790530408493829 2.8625045951371134e-06 -1.742456965739765e-05 2.272696183944069e-05
15 TSFA-main audio vision 10 100 37 0.24386757700414372 0.2190698984802626 0.2733444834273454 0.2444317758278318 0.21902110992468563 0.27219107996658004 0.000564198823688062 -0.00022641418153737327 0.001395855642402091
16 TSFA-main audio vision 2 100 37 0.24895100756913924 0.22416956525924045 0.27810367903911887 0.24896510513695155 0.22274126205659267 0.27930326741638245 1.4097567812305108e-05 -4.732088382019858e-05 7.63355802100572e-05
17 TSFA-main audio vision 5 100 37 0.2463452433058948 0.22097869036887005 0.2751587883843783 0.246497264087475 0.2214858425035993 0.2751850658218311 0.0001520207815802389 -0.00014739758679055951 0.0004554734628724275
18 TSFA-main text vision 1 100 37 0.3476741555274325 0.32148611245427566 0.3737497650390369 0.3476541439183967 0.32086054265937813 0.37359475283999777 -2.0011609035824965e-05 -6.729367620210594e-05 3.311202856174665e-05
19 TSFA-main text vision 10 100 37 0.33808821519346943 0.3123830481146549 0.36471129622609255 0.3382066619050689 0.31227041983330905 0.365546023769169 0.00011844671159944317 -0.001876128714584399 0.002146817830318111
20 TSFA-main text vision 2 100 37 0.3463484496062007 0.3194491144852671 0.372250999602138 0.34635784703948913 0.31847836298931187 0.3729820835754293 9.397433288411396e-06 -0.0001571169486153489 0.00020347758379682152
21 TSFA-main text vision 5 100 37 0.3427197299862511 0.31759720182711926 0.3687916805460675 0.34274824133142034 0.3161611675319563 0.37042456927433204 2.8511345169156534e-05 -0.0007572757773635113 0.0009358633216909969
22 TSFA-main vision audio 1 100 37 0.30385903824985366 0.2804930920345874 0.32447569376476193 0.3038967083492334 0.28237160721884863 0.32535285142726206 3.767009937983119e-05 2.2140309296097684e-05 5.2699988083822036e-05
23 TSFA-main vision audio 10 100 37 0.30352860244078117 0.2808613100727914 0.32450677675282535 0.30500827796376856 0.28273082537522215 0.32564273932073723 0.0014796755229874467 0.0004977929359841301 0.002395565682036881
24 TSFA-main vision audio 2 100 37 0.30382711134055096 0.2824193253098451 0.32531113133968326 0.3039569732806986 0.2827239277182888 0.3254230913166988 0.0001298619401475382 7.685713367700474e-05 0.00018430574637678292
25 TSFA-main vision audio 5 100 37 0.3037293283725082 0.2824902004786779 0.3239507917223273 0.30433064016570666 0.2829440499471156 0.32495928013412795 0.0006013117931984567 0.00028662674138598147 0.0009081425354786053
Binary file not shown.

After

Width:  |  Height:  |  Size: 170 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 86 KiB

@@ -0,0 +1,113 @@
method,modality,clip_count,video_id_count,span_ratio_video_macro_mean,span_ratio_ci95_low,span_ratio_ci95_high,span_abs_video_macro_mean,span_abs_ci95_low,span_abs_ci95_high,metric_family,direction,pairwise_time_mae_video_macro_mean,pairwise_time_mae_ci95_low,pairwise_time_mae_ci95_high,pairwise_signed_lag_video_macro_mean,pairwise_signed_lag_ci95_low,pairwise_signed_lag_ci95_high,pairwise_time_corr_video_macro_mean,pairwise_time_corr_ci95_low,pairwise_time_corr_ci95_high,near_similarity_offdiag_video_macro_mean,near_similarity_offdiag_ci95_low,near_similarity_offdiag_ci95_high,far_similarity_video_macro_mean,far_similarity_ci95_low,far_similarity_ci95_high,d_self_offdiag_video_macro_mean,d_self_offdiag_ci95_low,d_self_offdiag_ci95_high,gram_target_error_video_macro_mean,gram_target_error_ci95_low,gram_target_error_ci95_high,far_slot_leakage_video_macro_mean,far_slot_leakage_ci95_low,far_slot_leakage_ci95_high,cycle_kind,cycle_band_error_video_macro_mean,cycle_band_error_ci95_low,cycle_band_error_ci95_high,cycle_time_mae_video_macro_mean,cycle_time_mae_ci95_low,cycle_time_mae_ci95_high,triangle_relative_error_video_macro_mean,triangle_relative_error_ci95_low,triangle_relative_error_ci95_high,triangle_mean_absolute_residual_video_macro_mean,triangle_mean_absolute_residual_ci95_low,triangle_mean_absolute_residual_ci95_high
M3_noSourceTime,audio,100,37,0.006845225875929812,-0.0050480489482916426,0.018641673964870674,0.036477922386970923,0.029721606294995286,0.04310413792211531,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_noSourceTime,text,100,37,0.8675032330277417,0.8513314214973553,0.8824147079299696,0.8675032330277417,0.8522296196594468,0.8833766268951103,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_noSourceTime,vision,100,37,-0.05248904935502325,-0.09102026382990809,-0.017373356121759984,0.1361241436408572,0.10828269435313069,0.16937841433957945,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_sourceTime,audio,100,37,0.7079631044011514,0.6640514063782721,0.7384979763997631,0.7079631044011514,0.6651266517684566,0.7387978141351451,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_sourceTime,text,100,37,0.8675032330277417,0.8519241948813162,0.8843134111737262,0.8675032330277417,0.8520634954683572,0.8832864640856598,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_sourceTime,vision,100,37,0.7103498443247823,0.6786859055886948,0.7367456721643969,0.7103498443247823,0.677706011592382,0.7352975597625276,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M4_sourceTime,audio,100,37,0.8157060578090206,0.8144226218056944,0.8168588184847212,0.8157060578090206,0.8144152058853479,0.8168682832834657,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M4_sourceTime,text,100,37,0.7210220398065448,0.7029622253382732,0.7391635974741213,0.7210220398065448,0.7032309030429688,0.7388928438031057,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M4_sourceTime,vision,100,37,0.8217648772803375,0.819098820782984,0.8244210758168445,0.8217648772803375,0.8191228203610597,0.8243458564337164,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-global,audio,100,37,0.011351551106837957,-0.03307923351329353,0.054755991186068075,0.1361641932261882,0.1071832420028644,0.16555720151779119,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-global,text,100,37,0.7210220398065448,0.7028018689779558,0.7385341768148932,0.7210220398065448,0.7016156508853206,0.7388307750319667,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-global,vision,100,37,0.007610485834433587,-0.016687070936055162,0.033969306453111976,0.0831604315308393,0.06578630184541784,0.10186488890564342,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-main,audio,100,37,0.7959866389347422,0.7814527810816472,0.8097362828007407,0.7959866389347422,0.7818257680156309,0.8106330514335037,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-main,text,100,37,0.7210220398065448,0.7032168935165822,0.7400873256704725,0.7210220398065448,0.7028208226665803,0.7398312410824844,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-main,vision,100,37,0.8207314282427551,0.8150341158499098,0.8263598218647393,0.8207314282427551,0.8149187170040457,0.8264408487068886,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-multiply,audio,100,37,0.8238465478704826,0.8087581509483757,0.8369243381130251,0.8238465478704826,0.8084151479166035,0.8377465327122026,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-multiply,text,100,37,0.7210220398065448,0.7018140627887535,0.7388406501835107,0.7210220398065448,0.7017115274716906,0.7390402836448626,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-multiply,vision,100,37,0.8564870598082609,0.8515092554836746,0.8619992878873571,0.8564870598082609,0.8514594765786335,0.8620531933055623,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-random,audio,100,37,-0.0166169608453845,-0.0356561105662797,0.003911856542961215,0.3092861632918937,0.29694859626772774,0.32168873174102913,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-random,text,100,37,0.7210220398065448,0.702517714202256,0.7388815238925032,0.7210220398065448,0.7032355236466896,0.7387423622324837,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-random,vision,100,37,0.009951817230022326,-0.009544235701251413,0.02983924649004834,0.31530318162219934,0.30375650755698547,0.327376648433025,span,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_noSourceTime,,100,37,,,,,,,pairwise,audio_to_text,0.24798689239530688,0.24575595320867452,0.250478564603917,-0.00905109914646526,-0.021659838618577246,0.0043487629126405004,0.06358454760547783,0.023034305728283397,0.10249668543842441,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_noSourceTime,,100,37,,,,,,,pairwise,audio_to_vision,0.2557268510896316,0.2522416884193085,0.2599540490992459,0.0010860293038389595,-0.021838431492360325,0.02637708897886559,0.023136812147259593,-0.017164462319666285,0.06361670928884612,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_noSourceTime,,100,37,,,,,,,pairwise,text_to_audio,0.23844018512711035,0.231815399826556,0.24513403638360018,0.00032643349356500825,-0.014206368505950655,0.013788368566719469,0.09157069878863097,0.017082456940376654,0.16326818673274548,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_noSourceTime,,100,37,,,,,,,pairwise,text_to_vision,0.2595995219781619,0.24768831061652558,0.2721962396716379,0.003549698052312745,-0.018195794363028227,0.025729048351632686,-0.07469627228957339,-0.1645982737378071,0.010895163182870584,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_noSourceTime,,100,37,,,,,,,pairwise,vision_to_audio,0.2517105464205569,0.24997783849782754,0.25325740669487573,0.00011944104763818307,-0.003998172286235301,0.004314578789024335,0.09065033078927832,0.005952490324212469,0.18367249135769936,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_noSourceTime,,100,37,,,,,,,pairwise,vision_to_text,0.26965636750629723,0.26312577044158164,0.27726571774760705,-0.005324066992981435,-0.019810339686355333,0.010465554443999218,-0.051735984712035554,-0.14065674992330893,0.032703681806886056,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_sourceTime,,100,37,,,,,,,pairwise,audio_to_text,0.0871124186546691,0.07843402180952203,0.0960482575299671,-0.00818185656226965,-0.02501026774573674,0.009866064174329241,0.9621272440842052,0.9404591626453755,0.9767108597657944,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_sourceTime,,100,37,,,,,,,pairwise,audio_to_vision,0.07549793127843017,0.06847063238991467,0.08375030573884443,-0.005625651477412908,-0.00969390107684296,-0.0018935553076301747,0.976149538484469,0.9633859880459819,0.9838961024157739,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_sourceTime,,100,37,,,,,,,pairwise,text_to_audio,0.11115512644896658,0.09863427941596914,0.12356354925002973,0.0029040672010041033,-0.017689209501853456,0.021674971143182058,0.8664779183674878,0.8310981906498358,0.8947861184932466,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_sourceTime,,100,37,,,,,,,pairwise,text_to_vision,0.1095721907332496,0.09817313494579982,0.12168360601054318,-0.00021640865527353695,-0.020344818529633008,0.019194392157659432,0.8786640862668412,0.8555847476303776,0.9012911708379915,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_sourceTime,,100,37,,,,,,,pairwise,vision_to_audio,0.08402641543628109,0.07674065296156558,0.09217593659353024,-0.0002759945725429673,-0.005976672227240552,0.005817899813735201,0.9615264449317401,0.9381665699920014,0.9756067583306635,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_sourceTime,,100,37,,,,,,,pairwise,vision_to_text,0.0949450733631415,0.08614114589590695,0.10400518407272195,-0.006395282222322348,-0.023404822078732648,0.011666152940526454,0.9565651139127582,0.9392471240397936,0.9696111837015309,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M4_sourceTime,,100,37,,,,,,,pairwise,audio_to_text,0.05986829347375184,0.05422676746679547,0.06582581936684692,0.001716346997700615,-0.006062248610466559,0.010155625110385253,0.9896387509611086,0.987113812806792,0.9918754847735682,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M4_sourceTime,,100,37,,,,,,,pairwise,audio_to_vision,0.024761989554881157,0.023888416321772084,0.025694809281355088,0.0009678201964747196,0.0004121556321871045,0.0015657592778028952,0.9963711557037225,0.9957911008064857,0.9968302062855288,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M4_sourceTime,,100,37,,,,,,,pairwise,text_to_audio,0.08413479872440241,0.07637751970403586,0.09244831443618172,0.003073342587004113,-0.008817055347750192,0.01405819189112211,0.9277882717821058,0.9106142612324963,0.943464413917021,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M4_sourceTime,,100,37,,,,,,,pairwise,text_to_vision,0.08428864072945397,0.07631423853160876,0.09282317706135165,0.0030182972199566836,-0.008150982694603395,0.013742263037027585,0.9277943139436867,0.9109454276967438,0.9437667482725357,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M4_sourceTime,,100,37,,,,,,,pairwise,vision_to_audio,0.03125012554416389,0.030335602973254293,0.032274370405520454,0.001699291733416523,0.000989529421278144,0.002399587124229476,0.994063490418897,0.9934018419958887,0.9946277486801368,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M4_sourceTime,,100,37,,,,,,,pairwise,vision_to_text,0.06582797990544002,0.06015672140269782,0.07191811912272289,0.002534441565149494,-0.0049269372098783275,0.01014627322915136,0.9873438070809311,0.9842728666509312,0.9898634520953487,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-global,,100,37,,,,,,,pairwise,audio_to_text,0.26426354529126855,0.2595954058664512,0.26927950613867524,0.016063363039220284,0.0023853956672613154,0.030029997969078993,0.01186519338523276,-0.017863932995588257,0.04408118069862936,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-global,,100,37,,,,,,,pairwise,audio_to_vision,0.251902193365334,0.24851626175956615,0.25607406149155476,0.011592058619731964,-0.0020735190406534266,0.026210686820579737,-0.011882097548093922,-0.041437837608961185,0.02005575593397121,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-global,,100,37,,,,,,,pairwise,text_to_audio,0.23919675578879268,0.22823033229896159,0.2502279509927412,0.011978075655547017,-0.007856421287997644,0.03205505955258488,0.13953505174336778,0.013684665832117467,0.26550892119657277,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-global,,100,37,,,,,,,pairwise,text_to_vision,0.23858647233188462,0.2303998823983567,0.24728701813446996,0.016471399245608764,-0.0030107452002579448,0.03881277808578741,0.028802362948568784,-0.10411775420285496,0.16248498491336846,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-global,,100,37,,,,,,,pairwise,vision_to_audio,0.2600867141079052,0.2555181459039208,0.26489728316899913,0.008798799409372195,-0.009991486618197236,0.027522525432970055,-0.04216116616051461,-0.11934540642052503,0.0328652169183437,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-global,,100,37,,,,,,,pairwise,vision_to_text,0.25645549371974924,0.25181104298018353,0.2616800858478354,0.011963496297170268,0.0027002137949521223,0.02234805357364569,0.011657691672614899,-0.06382264351518231,0.08717482762128388,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-main,,100,37,,,,,,,pairwise,audio_to_text,0.06742660977259347,0.06276358021854823,0.07237029198906914,0.001769438513107734,-0.005499522451483993,0.009628460807482234,0.9780673585858292,0.9740539184484459,0.9812739256641996,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-main,,100,37,,,,,,,pairwise,audio_to_vision,0.04204494685473837,0.04060578185676881,0.04347158780261231,0.0015684376347839893,-0.0005385557212511506,0.0036400813468450794,0.9876396523633675,0.9866859120783338,0.9885968762026767,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-main,,100,37,,,,,,,pairwise,text_to_audio,0.08441610461747974,0.07641255077281847,0.09256938825420448,0.0024054198545138397,-0.009065747922210807,0.012980256698477306,0.9275963343341015,0.9108534196618289,0.9432700846219623,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-main,,100,37,,,,,,,pairwise,text_to_vision,0.08452495474559611,0.07623102571244669,0.09289985937525803,0.0033036427730361436,-0.007851205779410854,0.013378385207246357,0.9274092030944906,0.9101469934454218,0.9433551123562987,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-main,,100,37,,,,,,,pairwise,vision_to_audio,0.030076064676432743,0.02785290171629763,0.03252893344676687,0.0005874198264963244,-0.0019138561045836968,0.003188964794821949,0.9926644930991895,0.9917709342590185,0.9934552026921571,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-main,,100,37,,,,,,,pairwise,vision_to_text,0.05890817948197322,0.053470287745186154,0.06463501870845278,0.001918705659802123,-0.0056993804592977496,0.009731904017812175,0.9857566395421482,0.98161901062293,0.9889256100291415,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-multiply,,100,37,,,,,,,pairwise,audio_to_text,0.06643963790591051,0.06155519203761027,0.07146318366180746,0.0016541105648428527,-0.005800171780808754,0.0094397165252077,0.9792184259320681,0.9752197300750737,0.9826263203417981,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-multiply,,100,37,,,,,,,pairwise,audio_to_vision,0.03748529088182362,0.03602949972109203,0.03887249988224152,0.0014145159042255155,-0.0004971187343514189,0.003414844650505779,0.9895384201014192,0.9886004595988371,0.9904223364933273,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-multiply,,100,37,,,,,,,pairwise,text_to_audio,0.08316689271595411,0.07493328226138725,0.09161901581232872,0.002656665242589696,-0.008838504815762998,0.01322843925099298,0.9277193243312696,0.9105277835589112,0.9435921704549525,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-multiply,,100,37,,,,,,,pairwise,text_to_vision,0.08272346692478777,0.07413101830975553,0.09082178262359049,0.00355221780618286,-0.00780020790439905,0.013463799558736902,0.92770141066974,0.9112267680307472,0.9435642584766777,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-multiply,,100,37,,,,,,,pairwise,vision_to_audio,0.026516252496326696,0.024272904323796834,0.02923892906045038,0.0005202078371720159,-0.0017816156849158036,0.0028475703813473827,0.9943015227769918,0.9935560002859497,0.9950439734823314,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-multiply,,100,37,,,,,,,pairwise,vision_to_text,0.059111147729345934,0.05318454866186355,0.06473659759075691,0.001869284025221193,-0.0058491839333650285,0.009862095021189055,0.9861672927109927,0.9819146359888574,0.9894506176914498,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-random,,100,37,,,,,,,pairwise,audio_to_text,0.2715302012965951,0.2681423482921384,0.2751148151494631,0.013213615803127883,0.001544596804972213,0.024384888097449788,-0.02156884837448982,-0.043228793267016175,0.0008287115908261907,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-random,,100,37,,,,,,,pairwise,audio_to_vision,0.26677251834192656,0.2644589349599799,0.26927277649842407,-0.0009915105457357748,-0.0036349776141373887,0.0018582449001258766,0.002305975374338194,-0.016209359818672282,0.021096996250978824,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-random,,100,37,,,,,,,pairwise,text_to_audio,0.24253396559006382,0.23648577481667343,0.24879672462738361,0.004271181421459896,-0.008669847073590682,0.016266469979640173,0.001809260826441729,-0.024212496786697527,0.02977938535091217,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-random,,100,37,,,,,,,pairwise,text_to_vision,0.24279182915437733,0.23618153703657,0.24918830521619556,0.002916123269236995,-0.011716715868149562,0.015546720870855023,-0.025966406801151545,-0.054607945323011,0.003413669136643002,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-random,,100,37,,,,,,,pairwise,vision_to_audio,0.26196779815139154,0.25996107208940733,0.2642539732400678,0.0027559449435993182,3.7462547767353755e-05,0.005297637651498782,-0.0016721355157619025,-0.025000031524045015,0.019068474158866765,,,,,,,,,,,,,,,,,,,,,,,,,,,,
TSFA-random,,100,37,,,,,,,pairwise,vision_to_text,0.2619983087285825,0.26002152342407947,0.2640970253943566,0.009418709922206357,0.001987751724951219,0.01781982791398481,-0.02090627930113151,-0.04115157465871522,-0.0008479818764069004,,,,,,,,,,,,,,,,,,,,,,,,,,,,
M3_noSourceTime,audio,100,37,,,,,,,self,,,,,,,,,,,0.8602562808430486,0.8390017979472224,0.8791960507765548,0.6306394423686962,0.6065736211847845,0.6559103310590667,0.22961683847435244,0.212597593561371,0.24839158745125373,1.5433979842530003,1.4939613439855566,1.5980224666375722,0.5914922344267804,0.5863327383501546,0.5961485323484096,,,,,,,,,,,,,
M3_noSourceTime,text,100,37,,,,,,,self,,,,,,,,,,,0.500730224675529,0.44326381155953154,0.5546875932491161,0.0,0.0,0.0,0.500730224675529,0.44678261442106343,0.5536740821511082,0.7783799210361851,0.7588314988752327,0.7978312145643258,0.0,0.0,0.0,,,,,,,,,,,,,
M3_noSourceTime,vision,100,37,,,,,,,self,,,,,,,,,,,0.8370010596898583,0.8128242535668444,0.8586645729538069,0.5354454345010294,0.49812339029733943,0.5758168318539664,0.3015556251888289,0.26835387949396683,0.33611183886562884,1.3908956425610952,1.3204375307722693,1.4698184122554232,0.5652966473540847,0.5525497540528237,0.5764936815046576,,,,,,,,,,,,,
M3_sourceTime,audio,100,37,,,,,,,self,,,,,,,,,,,0.888331670085538,0.8783799815302573,0.8982361516689842,0.2859968780215105,0.2614315436352547,0.30991001303306515,0.6023347920640273,0.5718641086403762,0.629589423002457,1.1051953048433216,1.0611940801301807,1.1477056646407984,0.3858874333468643,0.36389107499473,0.4080694269041795,,,,,,,,,,,,,
M3_sourceTime,text,100,37,,,,,,,self,,,,,,,,,,,0.500730224675529,0.4425607231213978,0.5514131309491841,0.0,0.0,0.0,0.500730224675529,0.44289256370891295,0.553421139441445,0.7783799210361851,0.7596998963592567,0.7978403791798905,0.0,0.0,0.0,,,,,,,,,,,,,
M3_sourceTime,vision,100,37,,,,,,,self,,,,,,,,,,,0.8923011998859715,0.8829162316726851,0.90115625806856,0.30215556168475666,0.2766014128544894,0.32747475171660667,0.5901456382012148,0.5624948401289267,0.6188530100994244,1.113534676309553,1.072232565941515,1.1558847797411864,0.3964994452097854,0.37548961116022583,0.41767726895743396,,,,,,,,,,,,,
M4_sourceTime,audio,100,37,,,,,,,self,,,,,,,,,,,0.9830707859067331,0.983043585033989,0.9830990215661007,0.0692560064447752,0.06915230942710256,0.06936442043075314,0.9138147794619581,0.9137097396996392,0.9139154768791944,0.5438185058326855,0.5430943251047008,0.5445142540617243,0.12587317163943731,0.1257309976018284,0.12601617989449548,,,,,,,,,,,,,
M4_sourceTime,text,100,37,,,,,,,self,,,,,,,,,,,0.98463863452608,0.9837836830897783,0.9854255401052563,0.12195832206472494,0.10937889431437446,0.13486592677771062,0.8626803124613551,0.8505551729542101,0.8741516645446641,0.7332360007514397,0.7053039293689665,0.7605882309710607,0.19238998771979546,0.17759862017965225,0.2070004338350631,,,,,,,,,,,,,
M4_sourceTime,vision,100,37,,,,,,,self,,,,,,,,,,,0.9828657132228624,0.9827565207179908,0.9829821773464078,0.07301019914488216,0.07189453444841783,0.07421181716145536,0.9098555140779804,0.9086199818433717,0.9109538990663043,0.5426103369833182,0.5395012878989437,0.5462207981931749,0.13257047995822666,0.13090495318430864,0.13445453758276793,,,,,,,,,,,,,
TSFA-global,audio,100,37,,,,,,,self,,,,,,,,,,,0.9381095443274116,0.931651754282041,0.9441185921559399,0.2100733717905093,0.1826601718796808,0.23920233233183366,0.7280361725369022,0.7001579464958054,0.7535776749762602,0.863381418331564,0.8039491487313329,0.926914950609264,0.33487438012325216,0.31245989188931467,0.35649616064506184,,,,,,,,,,,,,
TSFA-global,text,100,37,,,,,,,self,,,,,,,,,,,0.98463863452608,0.983766490951091,0.9853981353359472,0.12195832206472494,0.10965717998549025,0.1347841332424256,0.8626803124613551,0.850372569552986,0.8740441549811607,0.7332360007514397,0.7066870479875449,0.7613386633835179,0.19238998771979546,0.17703831482671462,0.20823798629303045,,,,,,,,,,,,,
TSFA-global,vision,100,37,,,,,,,self,,,,,,,,,,,0.9889972492867095,0.9853616333200549,0.9921496089242945,0.7152139051917641,0.6681896821127071,0.7602635452078422,0.27378334409494554,0.23153592011200502,0.3169545411183176,1.807907897230697,1.7252640583459942,1.8897482017579965,0.5648694933764876,0.5505857506060692,0.5788789461690218,,,,,,,,,,,,,
TSFA-main,audio,100,37,,,,,,,self,,,,,,,,,,,0.8234864222497095,0.8147252263635052,0.831638166641138,0.0018924057590828816,0.0014796837371070474,0.002338841789812211,0.8215940164906266,0.8128654426624269,0.8293378196210572,0.4197792178264907,0.39931766382912903,0.44031334323802956,0.005687440577387501,0.004467740371657415,0.0070866126639524945,,,,,,,,,,,,,
TSFA-main,text,100,37,,,,,,,self,,,,,,,,,,,0.98463863452608,0.9837952570538889,0.9854168029916435,0.12195832206472494,0.11024324958022892,0.13513622841986347,0.8626803124613551,0.8502792958299866,0.8738112439977024,0.7332360007514397,0.7062337654689973,0.7604249710025476,0.19238998771979546,0.17820739335916275,0.20748713402377209,,,,,,,,,,,,,
TSFA-main,vision,100,37,,,,,,,self,,,,,,,,,,,0.8753932551323013,0.8700816518650041,0.8800146684598407,0.004439952313729791,0.003958120248637802,0.004918994027823828,0.8709533028185714,0.8654470668263632,0.875883832637767,0.2922236461705599,0.27221210954296104,0.31353420473606874,0.010609511873370787,0.009417318597629527,0.011965775154103995,,,,,,,,,,,,,
TSFA-multiply,audio,100,37,,,,,,,self,,,,,,,,,,,0.845473601064013,0.8371369589334942,0.8535547155860039,0.0011535688717004406,0.0008681368939363554,0.0014583059130755891,0.8443200321923128,0.8358434386626328,0.8519076978453295,0.4008834873616406,0.37780546574374,0.4226541568072099,0.0033878968038142357,0.0025584266488971052,0.0043033148150613335,,,,,,,,,,,,,
TSFA-multiply,text,100,37,,,,,,,self,,,,,,,,,,,0.98463863452608,0.9837051549803183,0.9854353569171195,0.12195832206472494,0.11040081311310136,0.13455153639535794,0.8626803124613551,0.8512955021521286,0.8742087339006359,0.7332360007514397,0.7059750518908112,0.7605916492760649,0.19238998771979546,0.17717071592148534,0.20736117762042403,,,,,,,,,,,,,
TSFA-multiply,vision,100,37,,,,,,,self,,,,,,,,,,,0.902764058566973,0.8974145599745504,0.9077503722502418,0.0022335280035617423,0.0019388695415776,0.0025267593271357434,0.9005305305634114,0.8948633558813284,0.9057063079847478,0.2410330598364526,0.2190705020750994,0.26486362144176023,0.005374162383074902,0.0045517751203657825,0.006298533655693178,,,,,,,,,,,,,
TSFA-random,audio,100,37,,,,,,,self,,,,,,,,,,,0.1702136025448653,0.16723226208912714,0.17300510804100797,0.050006548425651956,0.04463245283092054,0.05590187111703176,0.12020705411921334,0.11476586705860664,0.125197794987648,0.9000364572774586,0.8908052350066653,0.9108907750588845,0.3193311199680904,0.29964326945087166,0.34036697059575866,,,,,,,,,,,,,
TSFA-random,text,100,37,,,,,,,self,,,,,,,,,,,0.98463863452608,0.9837430297876442,0.9854105897288065,0.12195832206472494,0.11012758265837151,0.13410818320090137,0.8626803124613551,0.8498772919284686,0.874141656735737,0.7332360007514397,0.7050931335501408,0.7594350878347045,0.19238998771979546,0.17794175293831158,0.20700749691688433,,,,,,,,,,,,,
TSFA-random,vision,100,37,,,,,,,self,,,,,,,,,,,0.18680846851618726,0.18398392204894654,0.18986701006886691,0.16210455704600274,0.1565246182058625,0.16698550554870373,0.02470391147018457,0.020263875415987974,0.029383917404735463,1.0959556252675098,1.0855062865988658,1.1058215053916503,0.5504916063539527,0.5429613692763432,0.5567723296495888,,,,,,,,,,,,,
M3_noSourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_audio_vision_audio,0.9654507654379716,0.94782301334197,0.9844262926673437,0.24526603600206048,0.24460695902603002,0.24592368024056546,,,,,,
M3_noSourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_audio_text,0.8658500951793273,0.8596118509176524,0.8712751183264686,0.21988855319734216,0.21315919945463932,0.22707554891642193,,,,,,
M3_noSourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_vision_text,0.8312663326293602,0.8178681469090896,0.8442766436361115,0.20514106384912809,0.19733560193931374,0.2124968724218096,,,,,,
M3_noSourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,triangle_TAV,,,,,,,0.635751241203901,0.6076922664336832,0.6627118953375897,0.019152084617695322,0.016176198013098882,0.02205733569262106
M3_sourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_audio_vision_audio,0.8045496673993995,0.7971204875711047,0.8125584394999472,0.11888443345978468,0.10957197271767381,0.12804746974727776,,,,,,
M3_sourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_audio_text,0.7470532880949274,0.7350454921488248,0.7583312459337788,0.10782301608973462,0.09783257218838352,0.11888516673233186,,,,,,
M3_sourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_vision_text,0.7488389508139521,0.7375039579370523,0.7604845451779533,0.11137686125855244,0.10127015365499387,0.12215765169648239,,,,,,
M3_sourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,triangle_TAV,,,,,,,0.39458475984546676,0.3873522859378844,0.40157171964300187,0.01299650427840006,0.011530419632056874,0.014404423079089927
M4_sourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_audio_vision_audio,0.7194542968836412,0.7187292856075248,0.7202319735193489,0.04983903677260223,0.049023056283463394,0.050749148327950494,,,,,,
M4_sourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_audio_text,0.8651297946131749,0.8462946860695034,0.8852096860574188,0.09564003516962925,0.08768005125014286,0.10423717662834384,,,,,,
M4_sourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_vision_text,0.8642305913190416,0.8457638814878798,0.8844126405604584,0.09851611937621746,0.09029651463872992,0.10699334122957067,,,,,,
M4_sourceTime,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,triangle_TAV,,,,,,,0.2262284485325185,0.22148258912718663,0.23134585693643994,0.006954086200606416,0.006118734019699923,0.007763513689622475
TSFA-global,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_audio_vision_audio,1.7233751514850173,1.5320620802026454,1.910227031436825,0.2499189757629977,0.24813426973160263,0.2517913851678485,,,,,,
TSFA-global,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_audio_text,0.8828054688922468,0.8655570599393629,0.9002497028299116,0.16017333436663472,0.15020650011568096,0.17071839324736315,,,,,,
TSFA-global,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_vision_text,0.948073999704302,0.9353821128883112,0.961931287382611,0.22096955633926135,0.21213126171141727,0.22922736277343997,,,,,,
TSFA-global,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,triangle_TAV,,,,,,,0.14486850608413515,0.12897322570694134,0.16042111551767863,0.003349657125055823,0.0027169777735415804,0.003995757491955577
TSFA-main,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_audio_vision_audio,2.0587887072533735,1.883379123386877,2.239275068016061,0.04549742980449285,0.04396053159546751,0.04718275588930528,,,,,,
TSFA-main,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_audio_text,0.8415374909118359,0.8187267280346352,0.8647714332096803,0.08541476449405673,0.07766053427838494,0.09406202419279779,,,,,,
TSFA-main,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_vision_text,0.8473648847790903,0.8255998143909318,0.871333941141899,0.08711804315672145,0.07865623181447821,0.09561705139066572,,,,,,
TSFA-main,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,triangle_TAV,,,,,,,0.11295304219180526,0.10829628941263399,0.11777329791566588,0.003259473054729029,0.0028647321008819352,0.003665811502401078
TSFA-multiply,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_audio_vision_audio,2.067465096935341,1.8999584180770852,2.2458561639507786,0.04087040783248778,0.039240224738941,0.04256950418053263,,,,,,
TSFA-multiply,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_audio_text,0.8402509149510545,0.8179641691851386,0.8632553287603144,0.08511486218799579,0.0765977381321801,0.09348774126063883,,,,,,
TSFA-multiply,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_vision_text,0.8443214704351736,0.8227938391074004,0.8671506370987065,0.08644753408290066,0.07827521794931933,0.09534041757944388,,,,,,
TSFA-multiply,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,triangle_TAV,,,,,,,0.11133536067721364,0.10623922285694568,0.11667533159216026,0.0031566936483280374,0.002768192325193071,0.0035930796529320273
TSFA-random,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_audio_vision_audio,1.2376846358141187,1.1866995355978647,1.2929412415837394,0.22290692396408593,0.22159528393492708,0.2241383736708128,,,,,,
TSFA-random,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_audio_text,0.8726010103709779,0.856940323413578,0.8882453329903173,0.15424833033806157,0.14485775852974697,0.16349338810652073,,,,,,
TSFA-random,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,cycle_text_vision_text,0.9356607187725524,0.9246008385013482,0.9475927270482284,0.2136385162721102,0.20676204288370334,0.22075529878895422,,,,,,
TSFA-random,,100,37,,,,,,,cycle_triangle,,,,,,,,,,,,,,,,,,,,,,,,,,triangle_TAV,,,,,,,0.2284731831664083,0.21969593281342203,0.2373126331123121,0.005839563997881332,0.005085714025060491,0.006556360621932
1 method modality clip_count video_id_count span_ratio_video_macro_mean span_ratio_ci95_low span_ratio_ci95_high span_abs_video_macro_mean span_abs_ci95_low span_abs_ci95_high metric_family direction pairwise_time_mae_video_macro_mean pairwise_time_mae_ci95_low pairwise_time_mae_ci95_high pairwise_signed_lag_video_macro_mean pairwise_signed_lag_ci95_low pairwise_signed_lag_ci95_high pairwise_time_corr_video_macro_mean pairwise_time_corr_ci95_low pairwise_time_corr_ci95_high near_similarity_offdiag_video_macro_mean near_similarity_offdiag_ci95_low near_similarity_offdiag_ci95_high far_similarity_video_macro_mean far_similarity_ci95_low far_similarity_ci95_high d_self_offdiag_video_macro_mean d_self_offdiag_ci95_low d_self_offdiag_ci95_high gram_target_error_video_macro_mean gram_target_error_ci95_low gram_target_error_ci95_high far_slot_leakage_video_macro_mean far_slot_leakage_ci95_low far_slot_leakage_ci95_high cycle_kind cycle_band_error_video_macro_mean cycle_band_error_ci95_low cycle_band_error_ci95_high cycle_time_mae_video_macro_mean cycle_time_mae_ci95_low cycle_time_mae_ci95_high triangle_relative_error_video_macro_mean triangle_relative_error_ci95_low triangle_relative_error_ci95_high triangle_mean_absolute_residual_video_macro_mean triangle_mean_absolute_residual_ci95_low triangle_mean_absolute_residual_ci95_high
2 M3_noSourceTime audio 100 37 0.006845225875929812 -0.0050480489482916426 0.018641673964870674 0.036477922386970923 0.029721606294995286 0.04310413792211531 span
3 M3_noSourceTime text 100 37 0.8675032330277417 0.8513314214973553 0.8824147079299696 0.8675032330277417 0.8522296196594468 0.8833766268951103 span
4 M3_noSourceTime vision 100 37 -0.05248904935502325 -0.09102026382990809 -0.017373356121759984 0.1361241436408572 0.10828269435313069 0.16937841433957945 span
5 M3_sourceTime audio 100 37 0.7079631044011514 0.6640514063782721 0.7384979763997631 0.7079631044011514 0.6651266517684566 0.7387978141351451 span
6 M3_sourceTime text 100 37 0.8675032330277417 0.8519241948813162 0.8843134111737262 0.8675032330277417 0.8520634954683572 0.8832864640856598 span
7 M3_sourceTime vision 100 37 0.7103498443247823 0.6786859055886948 0.7367456721643969 0.7103498443247823 0.677706011592382 0.7352975597625276 span
8 M4_sourceTime audio 100 37 0.8157060578090206 0.8144226218056944 0.8168588184847212 0.8157060578090206 0.8144152058853479 0.8168682832834657 span
9 M4_sourceTime text 100 37 0.7210220398065448 0.7029622253382732 0.7391635974741213 0.7210220398065448 0.7032309030429688 0.7388928438031057 span
10 M4_sourceTime vision 100 37 0.8217648772803375 0.819098820782984 0.8244210758168445 0.8217648772803375 0.8191228203610597 0.8243458564337164 span
11 TSFA-global audio 100 37 0.011351551106837957 -0.03307923351329353 0.054755991186068075 0.1361641932261882 0.1071832420028644 0.16555720151779119 span
12 TSFA-global text 100 37 0.7210220398065448 0.7028018689779558 0.7385341768148932 0.7210220398065448 0.7016156508853206 0.7388307750319667 span
13 TSFA-global vision 100 37 0.007610485834433587 -0.016687070936055162 0.033969306453111976 0.0831604315308393 0.06578630184541784 0.10186488890564342 span
14 TSFA-main audio 100 37 0.7959866389347422 0.7814527810816472 0.8097362828007407 0.7959866389347422 0.7818257680156309 0.8106330514335037 span
15 TSFA-main text 100 37 0.7210220398065448 0.7032168935165822 0.7400873256704725 0.7210220398065448 0.7028208226665803 0.7398312410824844 span
16 TSFA-main vision 100 37 0.8207314282427551 0.8150341158499098 0.8263598218647393 0.8207314282427551 0.8149187170040457 0.8264408487068886 span
17 TSFA-multiply audio 100 37 0.8238465478704826 0.8087581509483757 0.8369243381130251 0.8238465478704826 0.8084151479166035 0.8377465327122026 span
18 TSFA-multiply text 100 37 0.7210220398065448 0.7018140627887535 0.7388406501835107 0.7210220398065448 0.7017115274716906 0.7390402836448626 span
19 TSFA-multiply vision 100 37 0.8564870598082609 0.8515092554836746 0.8619992878873571 0.8564870598082609 0.8514594765786335 0.8620531933055623 span
20 TSFA-random audio 100 37 -0.0166169608453845 -0.0356561105662797 0.003911856542961215 0.3092861632918937 0.29694859626772774 0.32168873174102913 span
21 TSFA-random text 100 37 0.7210220398065448 0.702517714202256 0.7388815238925032 0.7210220398065448 0.7032355236466896 0.7387423622324837 span
22 TSFA-random vision 100 37 0.009951817230022326 -0.009544235701251413 0.02983924649004834 0.31530318162219934 0.30375650755698547 0.327376648433025 span
23 M3_noSourceTime 100 37 pairwise audio_to_text 0.24798689239530688 0.24575595320867452 0.250478564603917 -0.00905109914646526 -0.021659838618577246 0.0043487629126405004 0.06358454760547783 0.023034305728283397 0.10249668543842441
24 M3_noSourceTime 100 37 pairwise audio_to_vision 0.2557268510896316 0.2522416884193085 0.2599540490992459 0.0010860293038389595 -0.021838431492360325 0.02637708897886559 0.023136812147259593 -0.017164462319666285 0.06361670928884612
25 M3_noSourceTime 100 37 pairwise text_to_audio 0.23844018512711035 0.231815399826556 0.24513403638360018 0.00032643349356500825 -0.014206368505950655 0.013788368566719469 0.09157069878863097 0.017082456940376654 0.16326818673274548
26 M3_noSourceTime 100 37 pairwise text_to_vision 0.2595995219781619 0.24768831061652558 0.2721962396716379 0.003549698052312745 -0.018195794363028227 0.025729048351632686 -0.07469627228957339 -0.1645982737378071 0.010895163182870584
27 M3_noSourceTime 100 37 pairwise vision_to_audio 0.2517105464205569 0.24997783849782754 0.25325740669487573 0.00011944104763818307 -0.003998172286235301 0.004314578789024335 0.09065033078927832 0.005952490324212469 0.18367249135769936
28 M3_noSourceTime 100 37 pairwise vision_to_text 0.26965636750629723 0.26312577044158164 0.27726571774760705 -0.005324066992981435 -0.019810339686355333 0.010465554443999218 -0.051735984712035554 -0.14065674992330893 0.032703681806886056
29 M3_sourceTime 100 37 pairwise audio_to_text 0.0871124186546691 0.07843402180952203 0.0960482575299671 -0.00818185656226965 -0.02501026774573674 0.009866064174329241 0.9621272440842052 0.9404591626453755 0.9767108597657944
30 M3_sourceTime 100 37 pairwise audio_to_vision 0.07549793127843017 0.06847063238991467 0.08375030573884443 -0.005625651477412908 -0.00969390107684296 -0.0018935553076301747 0.976149538484469 0.9633859880459819 0.9838961024157739
31 M3_sourceTime 100 37 pairwise text_to_audio 0.11115512644896658 0.09863427941596914 0.12356354925002973 0.0029040672010041033 -0.017689209501853456 0.021674971143182058 0.8664779183674878 0.8310981906498358 0.8947861184932466
32 M3_sourceTime 100 37 pairwise text_to_vision 0.1095721907332496 0.09817313494579982 0.12168360601054318 -0.00021640865527353695 -0.020344818529633008 0.019194392157659432 0.8786640862668412 0.8555847476303776 0.9012911708379915
33 M3_sourceTime 100 37 pairwise vision_to_audio 0.08402641543628109 0.07674065296156558 0.09217593659353024 -0.0002759945725429673 -0.005976672227240552 0.005817899813735201 0.9615264449317401 0.9381665699920014 0.9756067583306635
34 M3_sourceTime 100 37 pairwise vision_to_text 0.0949450733631415 0.08614114589590695 0.10400518407272195 -0.006395282222322348 -0.023404822078732648 0.011666152940526454 0.9565651139127582 0.9392471240397936 0.9696111837015309
35 M4_sourceTime 100 37 pairwise audio_to_text 0.05986829347375184 0.05422676746679547 0.06582581936684692 0.001716346997700615 -0.006062248610466559 0.010155625110385253 0.9896387509611086 0.987113812806792 0.9918754847735682
36 M4_sourceTime 100 37 pairwise audio_to_vision 0.024761989554881157 0.023888416321772084 0.025694809281355088 0.0009678201964747196 0.0004121556321871045 0.0015657592778028952 0.9963711557037225 0.9957911008064857 0.9968302062855288
37 M4_sourceTime 100 37 pairwise text_to_audio 0.08413479872440241 0.07637751970403586 0.09244831443618172 0.003073342587004113 -0.008817055347750192 0.01405819189112211 0.9277882717821058 0.9106142612324963 0.943464413917021
38 M4_sourceTime 100 37 pairwise text_to_vision 0.08428864072945397 0.07631423853160876 0.09282317706135165 0.0030182972199566836 -0.008150982694603395 0.013742263037027585 0.9277943139436867 0.9109454276967438 0.9437667482725357
39 M4_sourceTime 100 37 pairwise vision_to_audio 0.03125012554416389 0.030335602973254293 0.032274370405520454 0.001699291733416523 0.000989529421278144 0.002399587124229476 0.994063490418897 0.9934018419958887 0.9946277486801368
40 M4_sourceTime 100 37 pairwise vision_to_text 0.06582797990544002 0.06015672140269782 0.07191811912272289 0.002534441565149494 -0.0049269372098783275 0.01014627322915136 0.9873438070809311 0.9842728666509312 0.9898634520953487
41 TSFA-global 100 37 pairwise audio_to_text 0.26426354529126855 0.2595954058664512 0.26927950613867524 0.016063363039220284 0.0023853956672613154 0.030029997969078993 0.01186519338523276 -0.017863932995588257 0.04408118069862936
42 TSFA-global 100 37 pairwise audio_to_vision 0.251902193365334 0.24851626175956615 0.25607406149155476 0.011592058619731964 -0.0020735190406534266 0.026210686820579737 -0.011882097548093922 -0.041437837608961185 0.02005575593397121
43 TSFA-global 100 37 pairwise text_to_audio 0.23919675578879268 0.22823033229896159 0.2502279509927412 0.011978075655547017 -0.007856421287997644 0.03205505955258488 0.13953505174336778 0.013684665832117467 0.26550892119657277
44 TSFA-global 100 37 pairwise text_to_vision 0.23858647233188462 0.2303998823983567 0.24728701813446996 0.016471399245608764 -0.0030107452002579448 0.03881277808578741 0.028802362948568784 -0.10411775420285496 0.16248498491336846
45 TSFA-global 100 37 pairwise vision_to_audio 0.2600867141079052 0.2555181459039208 0.26489728316899913 0.008798799409372195 -0.009991486618197236 0.027522525432970055 -0.04216116616051461 -0.11934540642052503 0.0328652169183437
46 TSFA-global 100 37 pairwise vision_to_text 0.25645549371974924 0.25181104298018353 0.2616800858478354 0.011963496297170268 0.0027002137949521223 0.02234805357364569 0.011657691672614899 -0.06382264351518231 0.08717482762128388
47 TSFA-main 100 37 pairwise audio_to_text 0.06742660977259347 0.06276358021854823 0.07237029198906914 0.001769438513107734 -0.005499522451483993 0.009628460807482234 0.9780673585858292 0.9740539184484459 0.9812739256641996
48 TSFA-main 100 37 pairwise audio_to_vision 0.04204494685473837 0.04060578185676881 0.04347158780261231 0.0015684376347839893 -0.0005385557212511506 0.0036400813468450794 0.9876396523633675 0.9866859120783338 0.9885968762026767
49 TSFA-main 100 37 pairwise text_to_audio 0.08441610461747974 0.07641255077281847 0.09256938825420448 0.0024054198545138397 -0.009065747922210807 0.012980256698477306 0.9275963343341015 0.9108534196618289 0.9432700846219623
50 TSFA-main 100 37 pairwise text_to_vision 0.08452495474559611 0.07623102571244669 0.09289985937525803 0.0033036427730361436 -0.007851205779410854 0.013378385207246357 0.9274092030944906 0.9101469934454218 0.9433551123562987
51 TSFA-main 100 37 pairwise vision_to_audio 0.030076064676432743 0.02785290171629763 0.03252893344676687 0.0005874198264963244 -0.0019138561045836968 0.003188964794821949 0.9926644930991895 0.9917709342590185 0.9934552026921571
52 TSFA-main 100 37 pairwise vision_to_text 0.05890817948197322 0.053470287745186154 0.06463501870845278 0.001918705659802123 -0.0056993804592977496 0.009731904017812175 0.9857566395421482 0.98161901062293 0.9889256100291415
53 TSFA-multiply 100 37 pairwise audio_to_text 0.06643963790591051 0.06155519203761027 0.07146318366180746 0.0016541105648428527 -0.005800171780808754 0.0094397165252077 0.9792184259320681 0.9752197300750737 0.9826263203417981
54 TSFA-multiply 100 37 pairwise audio_to_vision 0.03748529088182362 0.03602949972109203 0.03887249988224152 0.0014145159042255155 -0.0004971187343514189 0.003414844650505779 0.9895384201014192 0.9886004595988371 0.9904223364933273
55 TSFA-multiply 100 37 pairwise text_to_audio 0.08316689271595411 0.07493328226138725 0.09161901581232872 0.002656665242589696 -0.008838504815762998 0.01322843925099298 0.9277193243312696 0.9105277835589112 0.9435921704549525
56 TSFA-multiply 100 37 pairwise text_to_vision 0.08272346692478777 0.07413101830975553 0.09082178262359049 0.00355221780618286 -0.00780020790439905 0.013463799558736902 0.92770141066974 0.9112267680307472 0.9435642584766777
57 TSFA-multiply 100 37 pairwise vision_to_audio 0.026516252496326696 0.024272904323796834 0.02923892906045038 0.0005202078371720159 -0.0017816156849158036 0.0028475703813473827 0.9943015227769918 0.9935560002859497 0.9950439734823314
58 TSFA-multiply 100 37 pairwise vision_to_text 0.059111147729345934 0.05318454866186355 0.06473659759075691 0.001869284025221193 -0.0058491839333650285 0.009862095021189055 0.9861672927109927 0.9819146359888574 0.9894506176914498
59 TSFA-random 100 37 pairwise audio_to_text 0.2715302012965951 0.2681423482921384 0.2751148151494631 0.013213615803127883 0.001544596804972213 0.024384888097449788 -0.02156884837448982 -0.043228793267016175 0.0008287115908261907
60 TSFA-random 100 37 pairwise audio_to_vision 0.26677251834192656 0.2644589349599799 0.26927277649842407 -0.0009915105457357748 -0.0036349776141373887 0.0018582449001258766 0.002305975374338194 -0.016209359818672282 0.021096996250978824
61 TSFA-random 100 37 pairwise text_to_audio 0.24253396559006382 0.23648577481667343 0.24879672462738361 0.004271181421459896 -0.008669847073590682 0.016266469979640173 0.001809260826441729 -0.024212496786697527 0.02977938535091217
62 TSFA-random 100 37 pairwise text_to_vision 0.24279182915437733 0.23618153703657 0.24918830521619556 0.002916123269236995 -0.011716715868149562 0.015546720870855023 -0.025966406801151545 -0.054607945323011 0.003413669136643002
63 TSFA-random 100 37 pairwise vision_to_audio 0.26196779815139154 0.25996107208940733 0.2642539732400678 0.0027559449435993182 3.7462547767353755e-05 0.005297637651498782 -0.0016721355157619025 -0.025000031524045015 0.019068474158866765
64 TSFA-random 100 37 pairwise vision_to_text 0.2619983087285825 0.26002152342407947 0.2640970253943566 0.009418709922206357 0.001987751724951219 0.01781982791398481 -0.02090627930113151 -0.04115157465871522 -0.0008479818764069004
65 M3_noSourceTime audio 100 37 self 0.8602562808430486 0.8390017979472224 0.8791960507765548 0.6306394423686962 0.6065736211847845 0.6559103310590667 0.22961683847435244 0.212597593561371 0.24839158745125373 1.5433979842530003 1.4939613439855566 1.5980224666375722 0.5914922344267804 0.5863327383501546 0.5961485323484096
66 M3_noSourceTime text 100 37 self 0.500730224675529 0.44326381155953154 0.5546875932491161 0.0 0.0 0.0 0.500730224675529 0.44678261442106343 0.5536740821511082 0.7783799210361851 0.7588314988752327 0.7978312145643258 0.0 0.0 0.0
67 M3_noSourceTime vision 100 37 self 0.8370010596898583 0.8128242535668444 0.8586645729538069 0.5354454345010294 0.49812339029733943 0.5758168318539664 0.3015556251888289 0.26835387949396683 0.33611183886562884 1.3908956425610952 1.3204375307722693 1.4698184122554232 0.5652966473540847 0.5525497540528237 0.5764936815046576
68 M3_sourceTime audio 100 37 self 0.888331670085538 0.8783799815302573 0.8982361516689842 0.2859968780215105 0.2614315436352547 0.30991001303306515 0.6023347920640273 0.5718641086403762 0.629589423002457 1.1051953048433216 1.0611940801301807 1.1477056646407984 0.3858874333468643 0.36389107499473 0.4080694269041795
69 M3_sourceTime text 100 37 self 0.500730224675529 0.4425607231213978 0.5514131309491841 0.0 0.0 0.0 0.500730224675529 0.44289256370891295 0.553421139441445 0.7783799210361851 0.7596998963592567 0.7978403791798905 0.0 0.0 0.0
70 M3_sourceTime vision 100 37 self 0.8923011998859715 0.8829162316726851 0.90115625806856 0.30215556168475666 0.2766014128544894 0.32747475171660667 0.5901456382012148 0.5624948401289267 0.6188530100994244 1.113534676309553 1.072232565941515 1.1558847797411864 0.3964994452097854 0.37548961116022583 0.41767726895743396
71 M4_sourceTime audio 100 37 self 0.9830707859067331 0.983043585033989 0.9830990215661007 0.0692560064447752 0.06915230942710256 0.06936442043075314 0.9138147794619581 0.9137097396996392 0.9139154768791944 0.5438185058326855 0.5430943251047008 0.5445142540617243 0.12587317163943731 0.1257309976018284 0.12601617989449548
72 M4_sourceTime text 100 37 self 0.98463863452608 0.9837836830897783 0.9854255401052563 0.12195832206472494 0.10937889431437446 0.13486592677771062 0.8626803124613551 0.8505551729542101 0.8741516645446641 0.7332360007514397 0.7053039293689665 0.7605882309710607 0.19238998771979546 0.17759862017965225 0.2070004338350631
73 M4_sourceTime vision 100 37 self 0.9828657132228624 0.9827565207179908 0.9829821773464078 0.07301019914488216 0.07189453444841783 0.07421181716145536 0.9098555140779804 0.9086199818433717 0.9109538990663043 0.5426103369833182 0.5395012878989437 0.5462207981931749 0.13257047995822666 0.13090495318430864 0.13445453758276793
74 TSFA-global audio 100 37 self 0.9381095443274116 0.931651754282041 0.9441185921559399 0.2100733717905093 0.1826601718796808 0.23920233233183366 0.7280361725369022 0.7001579464958054 0.7535776749762602 0.863381418331564 0.8039491487313329 0.926914950609264 0.33487438012325216 0.31245989188931467 0.35649616064506184
75 TSFA-global text 100 37 self 0.98463863452608 0.983766490951091 0.9853981353359472 0.12195832206472494 0.10965717998549025 0.1347841332424256 0.8626803124613551 0.850372569552986 0.8740441549811607 0.7332360007514397 0.7066870479875449 0.7613386633835179 0.19238998771979546 0.17703831482671462 0.20823798629303045
76 TSFA-global vision 100 37 self 0.9889972492867095 0.9853616333200549 0.9921496089242945 0.7152139051917641 0.6681896821127071 0.7602635452078422 0.27378334409494554 0.23153592011200502 0.3169545411183176 1.807907897230697 1.7252640583459942 1.8897482017579965 0.5648694933764876 0.5505857506060692 0.5788789461690218
77 TSFA-main audio 100 37 self 0.8234864222497095 0.8147252263635052 0.831638166641138 0.0018924057590828816 0.0014796837371070474 0.002338841789812211 0.8215940164906266 0.8128654426624269 0.8293378196210572 0.4197792178264907 0.39931766382912903 0.44031334323802956 0.005687440577387501 0.004467740371657415 0.0070866126639524945
78 TSFA-main text 100 37 self 0.98463863452608 0.9837952570538889 0.9854168029916435 0.12195832206472494 0.11024324958022892 0.13513622841986347 0.8626803124613551 0.8502792958299866 0.8738112439977024 0.7332360007514397 0.7062337654689973 0.7604249710025476 0.19238998771979546 0.17820739335916275 0.20748713402377209
79 TSFA-main vision 100 37 self 0.8753932551323013 0.8700816518650041 0.8800146684598407 0.004439952313729791 0.003958120248637802 0.004918994027823828 0.8709533028185714 0.8654470668263632 0.875883832637767 0.2922236461705599 0.27221210954296104 0.31353420473606874 0.010609511873370787 0.009417318597629527 0.011965775154103995
80 TSFA-multiply audio 100 37 self 0.845473601064013 0.8371369589334942 0.8535547155860039 0.0011535688717004406 0.0008681368939363554 0.0014583059130755891 0.8443200321923128 0.8358434386626328 0.8519076978453295 0.4008834873616406 0.37780546574374 0.4226541568072099 0.0033878968038142357 0.0025584266488971052 0.0043033148150613335
81 TSFA-multiply text 100 37 self 0.98463863452608 0.9837051549803183 0.9854353569171195 0.12195832206472494 0.11040081311310136 0.13455153639535794 0.8626803124613551 0.8512955021521286 0.8742087339006359 0.7332360007514397 0.7059750518908112 0.7605916492760649 0.19238998771979546 0.17717071592148534 0.20736117762042403
82 TSFA-multiply vision 100 37 self 0.902764058566973 0.8974145599745504 0.9077503722502418 0.0022335280035617423 0.0019388695415776 0.0025267593271357434 0.9005305305634114 0.8948633558813284 0.9057063079847478 0.2410330598364526 0.2190705020750994 0.26486362144176023 0.005374162383074902 0.0045517751203657825 0.006298533655693178
83 TSFA-random audio 100 37 self 0.1702136025448653 0.16723226208912714 0.17300510804100797 0.050006548425651956 0.04463245283092054 0.05590187111703176 0.12020705411921334 0.11476586705860664 0.125197794987648 0.9000364572774586 0.8908052350066653 0.9108907750588845 0.3193311199680904 0.29964326945087166 0.34036697059575866
84 TSFA-random text 100 37 self 0.98463863452608 0.9837430297876442 0.9854105897288065 0.12195832206472494 0.11012758265837151 0.13410818320090137 0.8626803124613551 0.8498772919284686 0.874141656735737 0.7332360007514397 0.7050931335501408 0.7594350878347045 0.19238998771979546 0.17794175293831158 0.20700749691688433
85 TSFA-random vision 100 37 self 0.18680846851618726 0.18398392204894654 0.18986701006886691 0.16210455704600274 0.1565246182058625 0.16698550554870373 0.02470391147018457 0.020263875415987974 0.029383917404735463 1.0959556252675098 1.0855062865988658 1.1058215053916503 0.5504916063539527 0.5429613692763432 0.5567723296495888
86 M3_noSourceTime 100 37 cycle_triangle cycle_audio_vision_audio 0.9654507654379716 0.94782301334197 0.9844262926673437 0.24526603600206048 0.24460695902603002 0.24592368024056546
87 M3_noSourceTime 100 37 cycle_triangle cycle_text_audio_text 0.8658500951793273 0.8596118509176524 0.8712751183264686 0.21988855319734216 0.21315919945463932 0.22707554891642193
88 M3_noSourceTime 100 37 cycle_triangle cycle_text_vision_text 0.8312663326293602 0.8178681469090896 0.8442766436361115 0.20514106384912809 0.19733560193931374 0.2124968724218096
89 M3_noSourceTime 100 37 cycle_triangle triangle_TAV 0.635751241203901 0.6076922664336832 0.6627118953375897 0.019152084617695322 0.016176198013098882 0.02205733569262106
90 M3_sourceTime 100 37 cycle_triangle cycle_audio_vision_audio 0.8045496673993995 0.7971204875711047 0.8125584394999472 0.11888443345978468 0.10957197271767381 0.12804746974727776
91 M3_sourceTime 100 37 cycle_triangle cycle_text_audio_text 0.7470532880949274 0.7350454921488248 0.7583312459337788 0.10782301608973462 0.09783257218838352 0.11888516673233186
92 M3_sourceTime 100 37 cycle_triangle cycle_text_vision_text 0.7488389508139521 0.7375039579370523 0.7604845451779533 0.11137686125855244 0.10127015365499387 0.12215765169648239
93 M3_sourceTime 100 37 cycle_triangle triangle_TAV 0.39458475984546676 0.3873522859378844 0.40157171964300187 0.01299650427840006 0.011530419632056874 0.014404423079089927
94 M4_sourceTime 100 37 cycle_triangle cycle_audio_vision_audio 0.7194542968836412 0.7187292856075248 0.7202319735193489 0.04983903677260223 0.049023056283463394 0.050749148327950494
95 M4_sourceTime 100 37 cycle_triangle cycle_text_audio_text 0.8651297946131749 0.8462946860695034 0.8852096860574188 0.09564003516962925 0.08768005125014286 0.10423717662834384
96 M4_sourceTime 100 37 cycle_triangle cycle_text_vision_text 0.8642305913190416 0.8457638814878798 0.8844126405604584 0.09851611937621746 0.09029651463872992 0.10699334122957067
97 M4_sourceTime 100 37 cycle_triangle triangle_TAV 0.2262284485325185 0.22148258912718663 0.23134585693643994 0.006954086200606416 0.006118734019699923 0.007763513689622475
98 TSFA-global 100 37 cycle_triangle cycle_audio_vision_audio 1.7233751514850173 1.5320620802026454 1.910227031436825 0.2499189757629977 0.24813426973160263 0.2517913851678485
99 TSFA-global 100 37 cycle_triangle cycle_text_audio_text 0.8828054688922468 0.8655570599393629 0.9002497028299116 0.16017333436663472 0.15020650011568096 0.17071839324736315
100 TSFA-global 100 37 cycle_triangle cycle_text_vision_text 0.948073999704302 0.9353821128883112 0.961931287382611 0.22096955633926135 0.21213126171141727 0.22922736277343997
101 TSFA-global 100 37 cycle_triangle triangle_TAV 0.14486850608413515 0.12897322570694134 0.16042111551767863 0.003349657125055823 0.0027169777735415804 0.003995757491955577
102 TSFA-main 100 37 cycle_triangle cycle_audio_vision_audio 2.0587887072533735 1.883379123386877 2.239275068016061 0.04549742980449285 0.04396053159546751 0.04718275588930528
103 TSFA-main 100 37 cycle_triangle cycle_text_audio_text 0.8415374909118359 0.8187267280346352 0.8647714332096803 0.08541476449405673 0.07766053427838494 0.09406202419279779
104 TSFA-main 100 37 cycle_triangle cycle_text_vision_text 0.8473648847790903 0.8255998143909318 0.871333941141899 0.08711804315672145 0.07865623181447821 0.09561705139066572
105 TSFA-main 100 37 cycle_triangle triangle_TAV 0.11295304219180526 0.10829628941263399 0.11777329791566588 0.003259473054729029 0.0028647321008819352 0.003665811502401078
106 TSFA-multiply 100 37 cycle_triangle cycle_audio_vision_audio 2.067465096935341 1.8999584180770852 2.2458561639507786 0.04087040783248778 0.039240224738941 0.04256950418053263
107 TSFA-multiply 100 37 cycle_triangle cycle_text_audio_text 0.8402509149510545 0.8179641691851386 0.8632553287603144 0.08511486218799579 0.0765977381321801 0.09348774126063883
108 TSFA-multiply 100 37 cycle_triangle cycle_text_vision_text 0.8443214704351736 0.8227938391074004 0.8671506370987065 0.08644753408290066 0.07827521794931933 0.09534041757944388
109 TSFA-multiply 100 37 cycle_triangle triangle_TAV 0.11133536067721364 0.10623922285694568 0.11667533159216026 0.0031566936483280374 0.002768192325193071 0.0035930796529320273
110 TSFA-random 100 37 cycle_triangle cycle_audio_vision_audio 1.2376846358141187 1.1866995355978647 1.2929412415837394 0.22290692396408593 0.22159528393492708 0.2241383736708128
111 TSFA-random 100 37 cycle_triangle cycle_text_audio_text 0.8726010103709779 0.856940323413578 0.8882453329903173 0.15424833033806157 0.14485775852974697 0.16349338810652073
112 TSFA-random 100 37 cycle_triangle cycle_text_vision_text 0.9356607187725524 0.9246008385013482 0.9475927270482284 0.2136385162721102 0.20676204288370334 0.22075529878895422
113 TSFA-random 100 37 cycle_triangle triangle_TAV 0.2284731831664083 0.21969593281342203 0.2373126331123121 0.005839563997881332 0.005085714025060491 0.006556360621932
Binary file not shown.

After

Width:  |  Height:  |  Size: 186 KiB

@@ -0,0 +1,43 @@
{
"seed": 42,
"grid_size": 50,
"hidden_size": 128,
"heads_in_frozen_alignment": 4,
"delta_normalized_time": 0.1,
"hard_negative_offsets_slots": [
-5,
-3,
-2,
2,
3,
5
],
"semantic_epochs": 40,
"semantic_learning_rate": 0.001,
"local_contrastive_temperature": 0.1,
"content_probe_epochs": 40,
"content_probe_temperature": 0.1,
"reconstruction_probe_epochs": 40,
"shuffle_repeats": 20,
"random_window_repeats": 20,
"window_ablation_status": "delta sweep and 80-percent attention-mass window deferred until the seven-method MVP is reviewed",
"semantic_query_key_inputs": "Frozen M4 attention-pooled text value output is the semantic query input; M4 pre-attention projected Audio/Vision content is the semantic key/value input. No explicit source-time code, absolute PE, slot ID, or latent positional vector is passed to semantic Q/K. Temporal selection can still encode time indirectly.",
"baseline_content_values": "M4 and M3 Audio/Vision use each checkpoint's actual MHA attention output. M3 Text uses attention-pooled pre-attention text features to remove its source-time query residual.",
"temporal_role": "Frozen M4 attention supplies candidate masks; TSFA-multiply additionally multiplies semantic probabilities by M4 temporal attention",
"local_contrastive": "same-slot positives with within-clip slot-offset negatives at +/-2, +/-3, +/-5; symmetric Text-Audio and Text-Vision loss",
"random_candidate": "same delta-width windows with independently sampled centers at inference; 20 held-out random draws, one train draw for the train-only probe",
"global_candidate": "all valid source positions available to semantic attention",
"content_probe": "same train-only symmetric within-clip InfoNCE probe applied to each method; held-out negatives are positions more than two slots away",
"content_shuffle": "independently permute each modality's 50 projected content rows within each held-out clip; M4 and TSFA-main",
"reconstruction": "fold-train decoder predicts one aligned content representation from the other two; shift one partner by signed offsets +/-1,2,5,10 on identical target support",
"confidence_intervals": "95-percent video_id-cluster bootstrap; 2,000 repetitions for CSV summaries",
"methods": [
"M3_noSourceTime",
"M3_sourceTime",
"M4_sourceTime",
"TSFA-main",
"TSFA-multiply",
"TSFA-random",
"TSFA-global"
]
}
@@ -0,0 +1,22 @@
method,modality,clip_count,video_id_count,mvr_epsilon_0_01_video_macro_mean,mvr_epsilon_0_01_ci95_low,mvr_epsilon_0_01_ci95_high,mvr_epsilon_0_02_video_macro_mean,mvr_epsilon_0_02_ci95_low,mvr_epsilon_0_02_ci95_high,mvr_epsilon_0_05_video_macro_mean,mvr_epsilon_0_05_ci95_low,mvr_epsilon_0_05_ci95_high,time_span_ratio_video_macro_mean,time_span_ratio_ci95_low,time_span_ratio_ci95_high,normalized_entropy_mean_video_macro_mean,normalized_entropy_mean_ci95_low,normalized_entropy_mean_ci95_high,source_coverage_rate_video_macro_mean,source_coverage_rate_ci95_low,source_coverage_rate_ci95_high
M3_noSourceTime,audio,100,37,0.1523511850042463,0.12890654138103116,0.1775852091158214,0.11118027648639893,0.09314737478002783,0.13030062861695513,0.03717244941734738,0.026531859848186373,0.04761878496572374,0.006845225875929802,-0.004974720338403389,0.018367859269390833,0.9484751192262098,0.9413206608520659,0.9551741392240333,1.0,1.0,1.0
M3_noSourceTime,text,100,37,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.8675032330277417,0.8521122687516515,0.8834943239261863,0.0013463322666157121,0.0,0.004038996799847136,1.0,1.0,1.0
M3_noSourceTime,vision,100,37,0.19206480533011147,0.16843243024365476,0.21904841795147914,0.1745917929591399,0.15144481215909786,0.1997835736611246,0.1288914278710197,0.10981094432625045,0.14901438245826,-0.05248904935502326,-0.09002182426718092,-0.018003616275419482,0.806956850982112,0.7777992190084679,0.8363112281176653,1.0,1.0,1.0
M3_sourceTime,audio,100,37,0.12775457673416857,0.10811011083970269,0.15060003151839885,0.11171039844509233,0.09435531741654192,0.1297671249967168,0.07801066372494944,0.06502721745068683,0.09146934835710345,0.7079631044011514,0.6658551697959926,0.7388692005361774,0.876785624402091,0.8714454313105274,0.8814951963640908,1.0,1.0,1.0
M3_sourceTime,text,100,37,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.8675032330277417,0.8514982524616859,0.8832306376250671,0.0013463322666157121,0.0,0.004038996799847136,1.0,1.0,1.0
M3_sourceTime,vision,100,37,0.12819977411814146,0.10739960076694771,0.152875036114832,0.11318563461420603,0.09480123709715546,0.13370256918726306,0.07447141018569589,0.062481876920652414,0.08859555984555983,0.7103498443247823,0.6770012181250491,0.7366984965947195,0.783255981912249,0.7721062307032261,0.7937545194100867,1.0,1.0,1.0
M4_sourceTime,audio,100,37,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.8157060578090207,0.8144300662672872,0.8168691650761236,0.8461043286957146,0.8432685890105509,0.8494121036930785,1.0,1.0,1.0
M4_sourceTime,text,100,37,0.0022181073201481363,0.0,0.005572452919391665,0.0003742809865258845,0.0,0.0009849499645418012,0.0,0.0,0.0,0.7210220398065448,0.7030382029586985,0.7392271515304444,0.6015413742024268,0.5867125625307114,0.6174784327556735,1.0,1.0,1.0
M4_sourceTime,vision,100,37,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.8217648772803375,0.8191443827978445,0.8242396484969566,0.7198475360014294,0.7103728027872257,0.7303326531663908,1.0,1.0,1.0
TSFA-global,audio,100,37,0.2408066959087367,0.21770717875309714,0.2635084793248058,0.14064997942548962,0.1231999054448034,0.157178588938793,0.03884773986814804,0.029588586399810893,0.04845730570220366,0.01135155110683782,-0.03307952783374511,0.05474875110063228,0.7009484166229463,0.6727335474417834,0.7300350355184457,1.0,1.0,1.0
TSFA-global,text,100,37,0.0022181073201481363,0.0,0.005677251595618942,0.0003742809865258845,0.0,0.0009465369159246656,0.0,0.0,0.0,0.7210220398065448,0.7023545769979086,0.7389899179294055,0.6015413742024268,0.5875561023863175,0.6159531085332476,1.0,1.0,1.0
TSFA-global,vision,100,37,0.13183358285399102,0.10105264885877131,0.16295599900191735,0.06452516656598291,0.043409021266164124,0.08763517891068912,0.017349565308748986,0.00888142515693536,0.027128378378378375,0.0076104858344335806,-0.01709729452706841,0.03299500925503682,0.902044345677258,0.8671364603137538,0.9334658138613803,1.0,1.0,1.0
TSFA-main,audio,100,37,0.01236571848816747,0.007674894282037139,0.018037126047330123,0.0025122790428912874,0.0009413495127780841,0.004597779266146612,0.0,0.0,0.0,0.7959866389347421,0.7821903087901576,0.8092532415645509,0.5376040991299903,0.5180895213701402,0.5563726279490699,0.9991265907079654,0.9981073190065879,0.9998740929454016
TSFA-main,text,100,37,0.0022181073201481363,0.0,0.005456622803561579,0.0003742809865258845,0.0,0.0009849499645418012,0.0,0.0,0.0,0.7210220398065448,0.7033838928902758,0.7389047563723402,0.6015413742024268,0.5877413970677293,0.6179347115827777,1.0,1.0,1.0
TSFA-main,vision,100,37,0.0015050035458198724,0.0005253066477556274,0.0027079557691802587,0.00038610038610038604,0.0,0.0010479867622724766,0.0,0.0,0.0,0.8207314282427551,0.8149307422602344,0.8265787064699319,0.5102220348192638,0.48883208000021033,0.5325813152434384,0.998100693277735,0.9965188082802303,0.9993628555025613
TSFA-multiply,audio,100,37,0.008198723504845955,0.004761904761904762,0.011943536101699364,0.002291650250833924,0.0006894649751792608,0.004294874320384522,0.0,0.0,0.0,0.8238465478704826,0.8092843768510277,0.8375431991484058,0.5374824091259056,0.5192627699574578,0.5557536982458872,0.9991146371356402,0.9980471424100044,0.9998737805676262
TSFA-multiply,text,100,37,0.0022181073201481363,0.0,0.00556693719959026,0.0003742809865258845,0.0,0.0009849499645418012,0.0,0.0,0.0,0.7210220398065448,0.7032826453493484,0.7394791749401468,0.6015413742024268,0.5869907574297073,0.6161966829904635,1.0,1.0,1.0
TSFA-multiply,vision,100,37,0.00011031439602868173,0.0,0.00033094318808604516,0.0,0.0,0.0,0.0,0.0,0.0,0.8564870598082609,0.8516115315250857,0.8619042260648179,0.5042486909404768,0.4827586923295134,0.5264450873938568,0.998100693277735,0.9964701640477626,0.9994824236000707
TSFA-random,audio,100,37,0.4774993433666905,0.47491765489979776,0.4801365665957502,0.46498947197926793,0.46206071997653625,0.4679666873506158,0.43554342972710314,0.43223226105113854,0.43878350143146055,-0.016616960845384495,-0.03615306587688328,0.0035000853492482787,0.5331781007342047,0.5159025794331448,0.5498651267241725,0.9998828757861041,0.9998175312333477,0.9999458535193156
TSFA-random,text,100,37,0.0022181073201481363,0.0,0.0058309037900874635,0.0003742809865258845,0.0,0.0009455519659601293,0.0,0.0,0.0,0.7210220398065448,0.7030686031936083,0.739263951757811,0.6015413742024268,0.5866909114370855,0.617601946399249,1.0,1.0,1.0
TSFA-random,vision,100,37,0.484447464957669,0.482029311016556,0.486984872810128,0.47450730613995923,0.4721397031141929,0.477074162901969,0.4430162451080818,0.440237471436451,0.4457687994116566,0.009951817230022326,-0.00961620625239858,0.030069627676334455,0.49993764529426016,0.47863484579793103,0.5198129423960058,0.9997262801373744,0.9995674285759633,0.9998587347587298
1 method modality clip_count video_id_count mvr_epsilon_0_01_video_macro_mean mvr_epsilon_0_01_ci95_low mvr_epsilon_0_01_ci95_high mvr_epsilon_0_02_video_macro_mean mvr_epsilon_0_02_ci95_low mvr_epsilon_0_02_ci95_high mvr_epsilon_0_05_video_macro_mean mvr_epsilon_0_05_ci95_low mvr_epsilon_0_05_ci95_high time_span_ratio_video_macro_mean time_span_ratio_ci95_low time_span_ratio_ci95_high normalized_entropy_mean_video_macro_mean normalized_entropy_mean_ci95_low normalized_entropy_mean_ci95_high source_coverage_rate_video_macro_mean source_coverage_rate_ci95_low source_coverage_rate_ci95_high
2 M3_noSourceTime audio 100 37 0.1523511850042463 0.12890654138103116 0.1775852091158214 0.11118027648639893 0.09314737478002783 0.13030062861695513 0.03717244941734738 0.026531859848186373 0.04761878496572374 0.006845225875929802 -0.004974720338403389 0.018367859269390833 0.9484751192262098 0.9413206608520659 0.9551741392240333 1.0 1.0 1.0
3 M3_noSourceTime text 100 37 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.8675032330277417 0.8521122687516515 0.8834943239261863 0.0013463322666157121 0.0 0.004038996799847136 1.0 1.0 1.0
4 M3_noSourceTime vision 100 37 0.19206480533011147 0.16843243024365476 0.21904841795147914 0.1745917929591399 0.15144481215909786 0.1997835736611246 0.1288914278710197 0.10981094432625045 0.14901438245826 -0.05248904935502326 -0.09002182426718092 -0.018003616275419482 0.806956850982112 0.7777992190084679 0.8363112281176653 1.0 1.0 1.0
5 M3_sourceTime audio 100 37 0.12775457673416857 0.10811011083970269 0.15060003151839885 0.11171039844509233 0.09435531741654192 0.1297671249967168 0.07801066372494944 0.06502721745068683 0.09146934835710345 0.7079631044011514 0.6658551697959926 0.7388692005361774 0.876785624402091 0.8714454313105274 0.8814951963640908 1.0 1.0 1.0
6 M3_sourceTime text 100 37 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.8675032330277417 0.8514982524616859 0.8832306376250671 0.0013463322666157121 0.0 0.004038996799847136 1.0 1.0 1.0
7 M3_sourceTime vision 100 37 0.12819977411814146 0.10739960076694771 0.152875036114832 0.11318563461420603 0.09480123709715546 0.13370256918726306 0.07447141018569589 0.062481876920652414 0.08859555984555983 0.7103498443247823 0.6770012181250491 0.7366984965947195 0.783255981912249 0.7721062307032261 0.7937545194100867 1.0 1.0 1.0
8 M4_sourceTime audio 100 37 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.8157060578090207 0.8144300662672872 0.8168691650761236 0.8461043286957146 0.8432685890105509 0.8494121036930785 1.0 1.0 1.0
9 M4_sourceTime text 100 37 0.0022181073201481363 0.0 0.005572452919391665 0.0003742809865258845 0.0 0.0009849499645418012 0.0 0.0 0.0 0.7210220398065448 0.7030382029586985 0.7392271515304444 0.6015413742024268 0.5867125625307114 0.6174784327556735 1.0 1.0 1.0
10 M4_sourceTime vision 100 37 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.8217648772803375 0.8191443827978445 0.8242396484969566 0.7198475360014294 0.7103728027872257 0.7303326531663908 1.0 1.0 1.0
11 TSFA-global audio 100 37 0.2408066959087367 0.21770717875309714 0.2635084793248058 0.14064997942548962 0.1231999054448034 0.157178588938793 0.03884773986814804 0.029588586399810893 0.04845730570220366 0.01135155110683782 -0.03307952783374511 0.05474875110063228 0.7009484166229463 0.6727335474417834 0.7300350355184457 1.0 1.0 1.0
12 TSFA-global text 100 37 0.0022181073201481363 0.0 0.005677251595618942 0.0003742809865258845 0.0 0.0009465369159246656 0.0 0.0 0.0 0.7210220398065448 0.7023545769979086 0.7389899179294055 0.6015413742024268 0.5875561023863175 0.6159531085332476 1.0 1.0 1.0
13 TSFA-global vision 100 37 0.13183358285399102 0.10105264885877131 0.16295599900191735 0.06452516656598291 0.043409021266164124 0.08763517891068912 0.017349565308748986 0.00888142515693536 0.027128378378378375 0.0076104858344335806 -0.01709729452706841 0.03299500925503682 0.902044345677258 0.8671364603137538 0.9334658138613803 1.0 1.0 1.0
14 TSFA-main audio 100 37 0.01236571848816747 0.007674894282037139 0.018037126047330123 0.0025122790428912874 0.0009413495127780841 0.004597779266146612 0.0 0.0 0.0 0.7959866389347421 0.7821903087901576 0.8092532415645509 0.5376040991299903 0.5180895213701402 0.5563726279490699 0.9991265907079654 0.9981073190065879 0.9998740929454016
15 TSFA-main text 100 37 0.0022181073201481363 0.0 0.005456622803561579 0.0003742809865258845 0.0 0.0009849499645418012 0.0 0.0 0.0 0.7210220398065448 0.7033838928902758 0.7389047563723402 0.6015413742024268 0.5877413970677293 0.6179347115827777 1.0 1.0 1.0
16 TSFA-main vision 100 37 0.0015050035458198724 0.0005253066477556274 0.0027079557691802587 0.00038610038610038604 0.0 0.0010479867622724766 0.0 0.0 0.0 0.8207314282427551 0.8149307422602344 0.8265787064699319 0.5102220348192638 0.48883208000021033 0.5325813152434384 0.998100693277735 0.9965188082802303 0.9993628555025613
17 TSFA-multiply audio 100 37 0.008198723504845955 0.004761904761904762 0.011943536101699364 0.002291650250833924 0.0006894649751792608 0.004294874320384522 0.0 0.0 0.0 0.8238465478704826 0.8092843768510277 0.8375431991484058 0.5374824091259056 0.5192627699574578 0.5557536982458872 0.9991146371356402 0.9980471424100044 0.9998737805676262
18 TSFA-multiply text 100 37 0.0022181073201481363 0.0 0.00556693719959026 0.0003742809865258845 0.0 0.0009849499645418012 0.0 0.0 0.0 0.7210220398065448 0.7032826453493484 0.7394791749401468 0.6015413742024268 0.5869907574297073 0.6161966829904635 1.0 1.0 1.0
19 TSFA-multiply vision 100 37 0.00011031439602868173 0.0 0.00033094318808604516 0.0 0.0 0.0 0.0 0.0 0.0 0.8564870598082609 0.8516115315250857 0.8619042260648179 0.5042486909404768 0.4827586923295134 0.5264450873938568 0.998100693277735 0.9964701640477626 0.9994824236000707
20 TSFA-random audio 100 37 0.4774993433666905 0.47491765489979776 0.4801365665957502 0.46498947197926793 0.46206071997653625 0.4679666873506158 0.43554342972710314 0.43223226105113854 0.43878350143146055 -0.016616960845384495 -0.03615306587688328 0.0035000853492482787 0.5331781007342047 0.5159025794331448 0.5498651267241725 0.9998828757861041 0.9998175312333477 0.9999458535193156
21 TSFA-random text 100 37 0.0022181073201481363 0.0 0.0058309037900874635 0.0003742809865258845 0.0 0.0009455519659601293 0.0 0.0 0.0 0.7210220398065448 0.7030686031936083 0.739263951757811 0.6015413742024268 0.5866909114370855 0.617601946399249 1.0 1.0 1.0
22 TSFA-random vision 100 37 0.484447464957669 0.482029311016556 0.486984872810128 0.47450730613995923 0.4721397031141929 0.477074162901969 0.4430162451080818 0.440237471436451 0.4457687994116566 0.009951817230022326 -0.00961620625239858 0.030069627676334455 0.49993764529426016 0.47863484579793103 0.5198129423960058 0.9997262801373744 0.9995674285759633 0.9998587347587298
Binary file not shown.

After

Width:  |  Height:  |  Size: 134 KiB