Complete standalone final deliverable and unaligned Q2 results

This commit is contained in:
2026-09-25 22:22:37 +08:00
parent c6b018e5d0
commit adc9c2064b
267 changed files with 15479 additions and 7976 deletions
+15
View File
@@ -0,0 +1,15 @@
model,family,split,n,accuracy,macro_f1,mae,rmse,pearson
C0,math,official_valid,728,0.5906593406593407,0.5573246854806906,0.7267872095108032,0.9493674330569902,0.5186371207237244
C1,math,official_valid,728,0.5879120879120879,0.47389937106918234,0.7446202039718628,0.9822674244990333,0.5362412333488464
C2,math,official_valid,728,0.5961538461538461,0.450273379924267,0.698158860206604,0.9260472748411018,0.5808078050613403
C3,math,official_valid,728,0.5906593406593407,0.443346825657295,0.7317898869514465,0.985695966827773,0.5456575155258179
C4,math,official_valid,728,0.554945054945055,0.4536351679548698,0.7454220056533813,1.0156130863371253,0.503315806388855
C5,math,official_valid,728,0.5824175824175825,0.4401015447285379,0.7787972092628479,1.024799634376836,0.5123178958892822
C6,math,official_valid,728,0.6085164835164835,0.518781225964892,0.6846672296524048,0.9208069125796768,0.6101511716842651
C6_no_distance,math,official_valid,728,0.6071428571428571,0.4720788453840183,0.7117781043052673,0.9322639795095579,0.6074535250663757
C6_no_reconstruction,math,official_valid,728,0.5837912087912088,0.48325805552048157,0.709170937538147,0.9543444540035635,0.5635251402854919
C6_pointmask,math,official_valid,728,0.5810439560439561,0.5025832785659426,0.679211437702179,0.9175777170595291,0.5958787798881531
C7_distill,math,official_valid,728,0.592032967032967,0.4870960207460408,0.7024686336517334,0.9442218762916826,0.5752816796302795
C7_group,math,official_valid,728,0.6002747252747253,0.4509009833805456,0.7323286533355713,0.9355743977553851,0.6008110046386719
EarlyConcat,deep_learning,official_valid,728,0.6236263736263736,0.5779072819396367,0.6614056825637817,0.8730861646443621,0.6092134453424162
MoFE-7,deep_learning,official_valid,728,0.614010989010989,0.5661110562085216,0.6426358222961426,0.8499269790699532,0.6013646920407029
1 model family split n accuracy macro_f1 mae rmse pearson
2 C0 math official_valid 728 0.5906593406593407 0.5573246854806906 0.7267872095108032 0.9493674330569902 0.5186371207237244
3 C1 math official_valid 728 0.5879120879120879 0.47389937106918234 0.7446202039718628 0.9822674244990333 0.5362412333488464
4 C2 math official_valid 728 0.5961538461538461 0.450273379924267 0.698158860206604 0.9260472748411018 0.5808078050613403
5 C3 math official_valid 728 0.5906593406593407 0.443346825657295 0.7317898869514465 0.985695966827773 0.5456575155258179
6 C4 math official_valid 728 0.554945054945055 0.4536351679548698 0.7454220056533813 1.0156130863371253 0.503315806388855
7 C5 math official_valid 728 0.5824175824175825 0.4401015447285379 0.7787972092628479 1.024799634376836 0.5123178958892822
8 C6 math official_valid 728 0.6085164835164835 0.518781225964892 0.6846672296524048 0.9208069125796768 0.6101511716842651
9 C6_no_distance math official_valid 728 0.6071428571428571 0.4720788453840183 0.7117781043052673 0.9322639795095579 0.6074535250663757
10 C6_no_reconstruction math official_valid 728 0.5837912087912088 0.48325805552048157 0.709170937538147 0.9543444540035635 0.5635251402854919
11 C6_pointmask math official_valid 728 0.5810439560439561 0.5025832785659426 0.679211437702179 0.9175777170595291 0.5958787798881531
12 C7_distill math official_valid 728 0.592032967032967 0.4870960207460408 0.7024686336517334 0.9442218762916826 0.5752816796302795
13 C7_group math official_valid 728 0.6002747252747253 0.4509009833805456 0.7323286533355713 0.9355743977553851 0.6008110046386719
14 EarlyConcat deep_learning official_valid 728 0.6236263736263736 0.5779072819396367 0.6614056825637817 0.8730861646443621 0.6092134453424162
15 MoFE-7 deep_learning official_valid 728 0.614010989010989 0.5661110562085216 0.6426358222961426 0.8499269790699532 0.6013646920407029