16 KiB
16 KiB
| 1 | model | rho_imp | lambda_u | lambda_gap | lambda_span | inner_selection_nll | scenario_selection_nll | selected | selection_split | note | group_lambda | group_temperature | risk_candidate_selected | candidate |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2 | teacher | 0.5 | 0.0 | 0.0 | 0.0 | 2.8498539328575134 | {"0.0/natural": 2.830216884613037, "0.3/single": 2.8259665966033936, "0.3/sync": 2.84348726272583, "0.5/async": 2.899744987487793} | False | reliability_validation | |||||
| 3 | teacher | 0.5 | 0.05 | 0.05 | 0.05 | 2.8493970036506653 | {"0.0/natural": 2.83003568649292, "0.3/single": 2.82570481300354, "0.3/sync": 2.8433682918548584, "0.5/async": 2.8984792232513428} | False | reliability_validation | |||||
| 4 | teacher | 0.5 | 0.1 | 0.0 | 0.0 | 2.8494723439216614 | {"0.0/natural": 2.830044984817505, "0.3/single": 2.825741767883301, "0.3/sync": 2.8434410095214844, "0.5/async": 2.8986616134643555} | False | reliability_validation | |||||
| 5 | teacher | 0.3 | 0.05 | 0.05 | 0.05 | 2.847354471683502 | {"0.0/natural": 2.82891583442688, "0.3/single": 2.8246304988861084, "0.3/sync": 2.8441710472106934, "0.5/async": 2.891700506210327} | True | reliability_validation | |||||
| 6 | teacher | 0.7 | 0.05 | 0.05 | 0.05 | 2.852314591407776 | {"0.0/natural": 2.8313441276550293, "0.3/single": 2.8275082111358643, "0.3/sync": 2.84439754486084, "0.5/async": 2.90600848197937} | False | reliability_validation | |||||
| 7 | C1 | 0.5 | 0.05 | 0.05 | 0.05 | nan | True | not used by this ablation | ||||||
| 8 | C2 | 0.5 | 0.05 | 0.05 | 0.05 | nan | True | not used by this ablation | ||||||
| 9 | C3 | 0.5 | 0.0 | 0.0 | 0.0 | 2.7700525522232056 | {"0.0/natural": 2.7679331302642822, "0.3/single": 2.766462802886963, "0.3/sync": 2.764604091644287, "0.5/async": 2.78121018409729} | False | reliability_validation | |||||
| 10 | C3 | 0.5 | 0.05 | 0.05 | 0.05 | 2.7690306305885315 | {"0.0/natural": 2.7670230865478516, "0.3/single": 2.7656285762786865, "0.3/sync": 2.763711452484131, "0.5/async": 2.779759407043457} | False | reliability_validation | |||||
| 11 | C3 | 0.5 | 0.1 | 0.0 | 0.0 | 2.7693098187446594 | {"0.0/natural": 2.7672297954559326, "0.3/single": 2.7659270763397217, "0.3/sync": 2.763989210128784, "0.5/async": 2.780093193054199} | False | reliability_validation | |||||
| 12 | C3 | 0.3 | 0.05 | 0.05 | 0.05 | 2.7640955448150635 | {"0.0/natural": 2.762216091156006, "0.3/single": 2.762289047241211, "0.3/sync": 2.7597885131835938, "0.5/async": 2.7720885276794434} | True | reliability_validation | |||||
| 13 | C3 | 0.7 | 0.05 | 0.05 | 0.05 | 2.7744842767715454 | {"0.0/natural": 2.7721493244171143, "0.3/single": 2.7696752548217773, "0.3/sync": 2.7683780193328857, "0.5/async": 2.7877345085144043} | False | reliability_validation | |||||
| 14 | C4 | 0.5 | 0.0 | 0.0 | 0.0 | 2.772478759288788 | {"0.0/natural": 2.754570484161377, "0.3/single": 2.750734567642212, "0.3/sync": 2.7757368087768555, "0.5/async": 2.808873176574707} | False | reliability_validation | |||||
| 15 | C4 | 0.5 | 0.05 | 0.05 | 0.05 | 2.772943437099457 | {"0.0/natural": 2.7551944255828857, "0.3/single": 2.7515814304351807, "0.3/sync": 2.7763662338256836, "0.5/async": 2.808631658554077} | False | reliability_validation | |||||
| 16 | C4 | 0.5 | 0.1 | 0.0 | 0.0 | 2.7727081179618835 | {"0.0/natural": 2.7549169063568115, "0.3/single": 2.751382350921631, "0.3/sync": 2.776087760925293, "0.5/async": 2.808445453643799} | False | reliability_validation | |||||
| 17 | C4 | 0.3 | 0.05 | 0.05 | 0.05 | 2.77541047334671 | {"0.0/natural": 2.7584164142608643, "0.3/single": 2.7576100826263428, "0.3/sync": 2.77994704246521, "0.5/async": 2.805668354034424} | False | reliability_validation | |||||
| 18 | C4 | 0.7 | 0.05 | 0.05 | 0.05 | 2.7717464566230774 | {"0.0/natural": 2.7531611919403076, "0.3/single": 2.7477409839630127, "0.3/sync": 2.774409294128418, "0.5/async": 2.8116743564605713} | True | reliability_validation | |||||
| 19 | C5 | 0.5 | 0.0 | 0.0 | 0.0 | 2.7821279764175415 | {"0.0/natural": 2.778216600418091, "0.3/single": 2.770395040512085, "0.3/sync": 2.7727770805358887, "0.5/async": 2.8071231842041016} | False | reliability_validation | |||||
| 20 | C5 | 0.5 | 0.05 | 0.05 | 0.05 | 2.7808529138565063 | {"0.0/natural": 2.7773375511169434, "0.3/single": 2.76949405670166, "0.3/sync": 2.7718558311462402, "0.5/async": 2.8047242164611816} | False | reliability_validation | |||||
| 21 | C5 | 0.5 | 0.1 | 0.0 | 0.0 | 2.781255841255188 | {"0.0/natural": 2.777573823928833, "0.3/single": 2.7698097229003906, "0.3/sync": 2.772275686264038, "0.5/async": 2.8053641319274902} | False | reliability_validation | |||||
| 22 | C5 | 0.3 | 0.05 | 0.05 | 0.05 | 2.7748592495918274 | {"0.0/natural": 2.772735834121704, "0.3/single": 2.76572847366333, "0.3/sync": 2.7685465812683105, "0.5/async": 2.792426109313965} | True | reliability_validation | |||||
| 23 | C5 | 0.7 | 0.05 | 0.05 | 0.05 | 2.7871546745300293 | {"0.0/natural": 2.781982183456421, "0.3/single": 2.7739250659942627, "0.3/sync": 2.77571177482605, "0.5/async": 2.816999673843384} | False | reliability_validation | |||||
| 24 | C6 | 0.5 | 0.0 | 0.0 | 0.0 | 2.8196974992752075 | {"0.0/natural": 2.8187146186828613, "0.3/single": 2.811687707901001, "0.3/sync": 2.824204206466675, "0.5/async": 2.824183464050293} | False | reliability_validation | |||||
| 25 | C6 | 0.5 | 0.05 | 0.05 | 0.05 | 2.818789839744568 | {"0.0/natural": 2.8179848194122314, "0.3/single": 2.811084032058716, "0.3/sync": 2.8237786293029785, "0.5/async": 2.8223118782043457} | False | reliability_validation | |||||
| 26 | C6 | 0.5 | 0.1 | 0.0 | 0.0 | 2.8189464807510376 | {"0.0/natural": 2.81801176071167, "0.3/single": 2.8111894130706787, "0.3/sync": 2.8239669799804688, "0.5/async": 2.822617769241333} | False | reliability_validation | |||||
| 27 | C6 | 0.3 | 0.05 | 0.05 | 0.05 | 2.8141549229621887 | {"0.0/natural": 2.813626289367676, "0.3/single": 2.8083877563476562, "0.3/sync": 2.823089838027954, "0.5/async": 2.8115158081054688} | True | reliability_validation | |||||
| 28 | C6 | 0.7 | 0.05 | 0.05 | 0.05 | 2.8246858716011047 | {"0.0/natural": 2.8233866691589355, "0.3/single": 2.815105676651001, "0.3/sync": 2.826352119445801, "0.5/async": 2.8338990211486816} | False | reliability_validation | |||||
| 29 | C6_no_distance | 0.5 | 0.0 | 0.0 | 0.0 | 2.8041199445724487 | {"0.0/natural": 2.8055496215820312, "0.3/single": 2.7918457984924316, "0.3/sync": 2.8010969161987305, "0.5/async": 2.8179874420166016} | True | reliability_validation | |||||
| 30 | C6_no_distance | 0.5 | 0.1 | 0.0 | 0.0 | 2.8041998147964478 | {"0.0/natural": 2.8057303428649902, "0.3/single": 2.792290687561035, "0.3/sync": 2.801347255706787, "0.5/async": 2.8174309730529785} | False | reliability_validation | |||||
| 31 | C6_no_reconstruction | 0.5 | 0.0 | 0.0 | 0.0 | 2.811096727848053 | {"0.0/natural": 2.81492280960083, "0.3/single": 2.802493095397949, "0.3/sync": 2.8050832748413086, "0.5/async": 2.821887731552124} | False | reliability_validation | |||||
| 32 | C6_no_reconstruction | 0.5 | 0.05 | 0.05 | 0.05 | 2.8102840781211853 | {"0.0/natural": 2.814619541168213, "0.3/single": 2.8024234771728516, "0.3/sync": 2.8045318126678467, "0.5/async": 2.81956148147583} | False | reliability_validation | |||||
| 33 | C6_no_reconstruction | 0.5 | 0.1 | 0.0 | 0.0 | 2.8104811310768127 | {"0.0/natural": 2.8146867752075195, "0.3/single": 2.802565097808838, "0.3/sync": 2.80460786819458, "0.5/async": 2.8200647830963135} | False | reliability_validation | |||||
| 34 | C6_no_reconstruction | 0.3 | 0.05 | 0.05 | 0.05 | 2.8067832589149475 | {"0.0/natural": 2.8130786418914795, "0.3/single": 2.8034427165985107, "0.3/sync": 2.8017995357513428, "0.5/async": 2.808812141418457} | True | reliability_validation | |||||
| 35 | C6_no_reconstruction | 0.7 | 0.05 | 0.05 | 0.05 | 2.814980149269104 | {"0.0/natural": 2.8163907527923584, "0.3/single": 2.802222967147827, "0.3/sync": 2.8083655834198, "0.5/async": 2.8329412937164307} | False | reliability_validation | |||||
| 36 | C6_pointmask | 0.5 | 0.0 | 0.0 | 0.0 | 2.8015413880348206 | {"0.0/natural": 2.795191764831543, "0.3/single": 2.784855842590332, "0.3/sync": 2.790245294570923, "0.5/async": 2.8358726501464844} | False | reliability_validation | |||||
| 37 | C6_pointmask | 0.5 | 0.05 | 0.05 | 0.05 | 2.800622880458832 | {"0.0/natural": 2.7952637672424316, "0.3/single": 2.785073757171631, "0.3/sync": 2.7895565032958984, "0.5/async": 2.832597494125366} | False | reliability_validation | |||||
| 38 | C6_pointmask | 0.5 | 0.1 | 0.0 | 0.0 | 2.8006376028060913 | {"0.0/natural": 2.795201301574707, "0.3/single": 2.7850871086120605, "0.3/sync": 2.789341449737549, "0.5/async": 2.832920551300049} | False | reliability_validation | |||||
| 39 | C6_pointmask | 0.3 | 0.05 | 0.05 | 0.05 | 2.7950855493545532 | {"0.0/natural": 2.7960610389709473, "0.3/single": 2.78786039352417, "0.3/sync": 2.7846107482910156, "0.5/async": 2.81181001663208} | True | reliability_validation | |||||
| 40 | C6_pointmask | 0.7 | 0.05 | 0.05 | 0.05 | 2.8071449995040894 | {"0.0/natural": 2.7956111431121826, "0.3/single": 2.784064292907715, "0.3/sync": 2.796639919281006, "0.5/async": 2.852264642715454} | False | reliability_validation | |||||
| 41 | C7_distill | 0.5 | 0.0 | 0.0 | 0.0 | 2.7951783537864685 | {"0.0/natural": 2.7921783924102783, "0.3/single": 2.7866597175598145, "0.3/sync": 2.799243688583374, "0.5/async": 2.8026316165924072} | False | reliability_validation | |||||
| 42 | C7_distill | 0.5 | 0.05 | 0.05 | 0.05 | 2.7945213317871094 | {"0.0/natural": 2.791801929473877, "0.3/single": 2.7864341735839844, "0.3/sync": 2.7990593910217285, "0.5/async": 2.8007898330688477} | False | reliability_validation | |||||
| 43 | C7_distill | 0.5 | 0.1 | 0.0 | 0.0 | 2.794647455215454 | {"0.0/natural": 2.791788101196289, "0.3/single": 2.7865233421325684, "0.3/sync": 2.799224615097046, "0.5/async": 2.801053762435913} | False | reliability_validation | |||||
| 44 | C7_distill | 0.3 | 0.05 | 0.05 | 0.05 | 2.7921347618103027 | {"0.0/natural": 2.7898409366607666, "0.3/single": 2.7869362831115723, "0.3/sync": 2.80041241645813, "0.5/async": 2.791349411010742} | True | reliability_validation | |||||
| 45 | C7_distill | 0.7 | 0.05 | 0.05 | 0.05 | 2.799198091030121 | {"0.0/natural": 2.795092821121216, "0.3/single": 2.7883641719818115, "0.3/sync": 2.8003005981445312, "0.5/async": 2.813034772872925} | False | reliability_validation | |||||
| 46 | C7_group | 0.5 | 0.0 | 0.0 | 0.0 | 2.828823149204254 | {"0.0/natural": 2.8187339305877686, "0.3/single": 2.8182389736175537, "0.3/sync": 2.8243916034698486, "0.5/async": 2.8539280891418457} | False | reliability_validation | 0.05 | 0.1 | False | C7_group_lambda0.05_tau0.10 | |
| 47 | C7_group | 0.5 | 0.05 | 0.05 | 0.05 | 2.828660488128662 | {"0.0/natural": 2.819159984588623, "0.3/single": 2.819512367248535, "0.3/sync": 2.8247179985046387, "0.5/async": 2.8512516021728516} | False | reliability_validation | 0.05 | 0.1 | False | C7_group_lambda0.05_tau0.10 | |
| 48 | C7_group | 0.5 | 0.1 | 0.0 | 0.0 | 2.8285528421401978 | {"0.0/natural": 2.8189356327056885, "0.3/single": 2.81935453414917, "0.3/sync": 2.824479579925537, "0.5/async": 2.8514416217803955} | False | reliability_validation | 0.05 | 0.1 | False | C7_group_lambda0.05_tau0.10 | |
| 49 | C7_group | 0.3 | 0.05 | 0.05 | 0.05 | 2.8293744921684265 | {"0.0/natural": 2.822007417678833, "0.3/single": 2.8294715881347656, "0.3/sync": 2.82853627204895, "0.5/async": 2.8374826908111572} | False | reliability_validation | 0.05 | 0.1 | False | C7_group_lambda0.05_tau0.10 | |
| 50 | C7_group | 0.7 | 0.05 | 0.05 | 0.05 | 2.8315007090568542 | {"0.0/natural": 2.818368911743164, "0.3/single": 2.8128044605255127, "0.3/sync": 2.825650930404663, "0.5/async": 2.869178533554077} | False | reliability_validation | 0.05 | 0.1 | False | C7_group_lambda0.05_tau0.10 | |
| 51 | C7_group | 0.5 | 0.0 | 0.0 | 0.0 | 2.779416024684906 | {"0.0/natural": 2.770010471343994, "0.3/single": 2.762016773223877, "0.3/sync": 2.7750518321990967, "0.5/async": 2.8105850219726562} | False | reliability_validation | 0.1 | 0.05 | True | C7_group_lambda0.10_tau0.05 | |
| 52 | C7_group | 0.5 | 0.05 | 0.05 | 0.05 | 2.777971565723419 | {"0.0/natural": 2.7691972255706787, "0.3/single": 2.761157512664795, "0.3/sync": 2.7739386558532715, "0.5/async": 2.8075928688049316} | False | reliability_validation | 0.1 | 0.05 | True | C7_group_lambda0.10_tau0.05 | |
| 53 | C7_group | 0.5 | 0.1 | 0.0 | 0.0 | 2.778379499912262 | {"0.0/natural": 2.7693538665771484, "0.3/single": 2.7614493370056152, "0.3/sync": 2.7743921279907227, "0.5/async": 2.8083226680755615} | False | reliability_validation | 0.1 | 0.05 | True | C7_group_lambda0.10_tau0.05 | |
| 54 | C7_group | 0.3 | 0.05 | 0.05 | 0.05 | 2.7721481323242188 | {"0.0/natural": 2.765040636062622, "0.3/single": 2.75869083404541, "0.3/sync": 2.7706310749053955, "0.5/async": 2.7942299842834473} | True | reliability_validation | 0.1 | 0.05 | True | C7_group_lambda0.10_tau0.05 | |
| 55 | C7_group | 0.7 | 0.05 | 0.05 | 0.05 | 2.785994231700897 | {"0.0/natural": 2.774202346801758, "0.3/single": 2.7659974098205566, "0.3/sync": 2.779461145401001, "0.5/async": 2.8243160247802734} | False | reliability_validation | 0.1 | 0.05 | True | C7_group_lambda0.10_tau0.05 | |
| 56 | C7_group | 0.5 | 0.0 | 0.0 | 0.0 | 2.789133071899414 | {"0.0/natural": 2.7846412658691406, "0.3/single": 2.7780208587646484, "0.3/sync": 2.7788124084472656, "0.5/async": 2.8150577545166016} | False | reliability_validation | 0.1 | 0.1 | False | C7_group_lambda0.10_tau0.10 | |
| 57 | C7_group | 0.5 | 0.05 | 0.05 | 0.05 | 2.78842294216156 | {"0.0/natural": 2.784080743789673, "0.3/single": 2.777449131011963, "0.3/sync": 2.77826189994812, "0.5/async": 2.8138999938964844} | False | reliability_validation | 0.1 | 0.1 | False | C7_group_lambda0.10_tau0.10 | |
| 58 | C7_group | 0.5 | 0.1 | 0.0 | 0.0 | 2.788659632205963 | {"0.0/natural": 2.784231662750244, "0.3/single": 2.777684211730957, "0.3/sync": 2.77851939201355, "0.5/async": 2.8142032623291016} | False | reliability_validation | 0.1 | 0.1 | False | C7_group_lambda0.10_tau0.10 | |
| 59 | C7_group | 0.3 | 0.05 | 0.05 | 0.05 | 2.7850821018218994 | {"0.0/natural": 2.7810800075531006, "0.3/single": 2.775282621383667, "0.3/sync": 2.775895357131958, "0.5/async": 2.808070421218872} | False | reliability_validation | 0.1 | 0.1 | False | C7_group_lambda0.10_tau0.10 | |
| 60 | C7_group | 0.7 | 0.05 | 0.05 | 0.05 | 2.7917492389678955 | {"0.0/natural": 2.7869956493377686, "0.3/single": 2.779932975769043, "0.3/sync": 2.7802255153656006, "0.5/async": 2.81984281539917} | False | reliability_validation | 0.1 | 0.1 | False | C7_group_lambda0.10_tau0.10 | |
| 61 | C7_group | 0.5 | 0.0 | 0.0 | 0.0 | 2.7906041741371155 | {"0.0/natural": 2.773076057434082, "0.3/single": 2.779714584350586, "0.3/sync": 2.779757261276245, "0.5/async": 2.829868793487549} | False | reliability_validation | 0.1 | 0.2 | False | C7_group_lambda0.10_tau0.20 | |
| 62 | C7_group | 0.5 | 0.05 | 0.05 | 0.05 | 2.7901496291160583 | {"0.0/natural": 2.773569107055664, "0.3/single": 2.780526876449585, "0.3/sync": 2.77995228767395, "0.5/async": 2.826550245285034} | False | reliability_validation | 0.1 | 0.2 | False | C7_group_lambda0.10_tau0.20 | |
| 63 | C7_group | 0.5 | 0.1 | 0.0 | 0.0 | 2.789774775505066 | {"0.0/natural": 2.773125648498535, "0.3/single": 2.780186176300049, "0.3/sync": 2.77935528755188, "0.5/async": 2.8264319896698} | False | reliability_validation | 0.1 | 0.2 | False | C7_group_lambda0.10_tau0.20 | |
| 64 | C7_group | 0.3 | 0.05 | 0.05 | 0.05 | 2.7867594957351685 | {"0.0/natural": 2.776130437850952, "0.3/single": 2.7867321968078613, "0.3/sync": 2.7799689769744873, "0.5/async": 2.804206371307373} | False | reliability_validation | 0.1 | 0.2 | False | C7_group_lambda0.10_tau0.20 | |
| 65 | C7_group | 0.7 | 0.05 | 0.05 | 0.05 | 2.797126829624176 | {"0.0/natural": 2.7745771408081055, "0.3/single": 2.7787585258483887, "0.3/sync": 2.7843544483184814, "0.5/async": 2.8508172035217285} | False | reliability_validation | 0.1 | 0.2 | False | C7_group_lambda0.10_tau0.20 | |
| 66 | C7_group | 0.5 | 0.0 | 0.0 | 0.0 | 2.82554292678833 | {"0.0/natural": 2.8007912635803223, "0.3/single": 2.79516339302063, "0.3/sync": 2.8298346996307373, "0.5/async": 2.876382350921631} | False | reliability_validation | 0.2 | 0.1 | False | C7_group_lambda0.20_tau0.10 | |
| 67 | C7_group | 0.5 | 0.05 | 0.05 | 0.05 | 2.824092984199524 | {"0.0/natural": 2.7999930381774902, "0.3/single": 2.794499158859253, "0.3/sync": 2.828606128692627, "0.5/async": 2.8732736110687256} | False | reliability_validation | 0.2 | 0.1 | False | C7_group_lambda0.20_tau0.10 | |
| 68 | C7_group | 0.5 | 0.1 | 0.0 | 0.0 | 2.8244409561157227 | {"0.0/natural": 2.80021595954895, "0.3/single": 2.79475736618042, "0.3/sync": 2.8288345336914062, "0.5/async": 2.8739559650421143} | False | reliability_validation | 0.2 | 0.1 | False | C7_group_lambda0.20_tau0.10 | |
| 69 | C7_group | 0.3 | 0.05 | 0.05 | 0.05 | 2.816759705543518 | {"0.0/natural": 2.7960336208343506, "0.3/single": 2.792614221572876, "0.3/sync": 2.8221685886383057, "0.5/async": 2.85622239112854} | False | reliability_validation | 0.2 | 0.1 | False | C7_group_lambda0.20_tau0.10 | |
| 70 | C7_group | 0.7 | 0.05 | 0.05 | 0.05 | 2.831870377063751 | {"0.0/natural": 2.8041059970855713, "0.3/single": 2.7978010177612305, "0.3/sync": 2.835824489593506, "0.5/async": 2.8897500038146973} | False | reliability_validation | 0.2 | 0.1 | False | C7_group_lambda0.20_tau0.10 |