Common Sense Reasoning on ARC-c, ARC-e, HS, LAMB, OBQA, PIQA, SciQ, Wino
25.13ARC-c AccuracySTR
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| STRG1=true, G2=true, G3=false, G4=true2026.05 | 25.13 | 37.17 | 28.13 | 12.25 | 27.3 | 56.07 | 57.3 | 52 | 36.92 | |
| SAWG1=true, G2=true, G3=true, G4=true2026.05 | 24.49 | 37.34 | 28.55 | 12.28 | 27.2 | 56.04 | 56.4 | 52.14 | 36.81 | |
| CLG1=true, G2=false, G3=true, G4=false2026.05 | 23.63 | 36.99 | 27.63 | 9.9 | 27 | 55.22 | 58.4 | 50.2 | 36.12 | |
| DELTG1=true, G2=true, G3=false, G4=false2026.05 | 23.29 | 36.57 | 27.96 | 11.14 | 27 | 56.04 | 59.6 | 51.38 | 36.62 | |
| RandomG1=false, G2=true, G3=false, G4=true2026.05 | 21.08 | 33.63 | 27.5 | 14.26 | 25.8 | 55.66 | 55.5 | 51.38 | 35.6 |