Question Answering on 2-Hop QA Parametric Knowledge
43.07AccuracyCoT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CoTBase Model=Qwen2.5-Base-1.5B2026.02 | 43.07 | 1 | |
| Thinking StatesBase Model=Qwen2.5-Base-1.5B2026.02 | 43.05 | 1.23 | |
| iCoTBase Model=Qwen2.5-Base-1.5B2026.02 | 36.31 | 1.8 | |
| CoconutBase Model=Qwen2.5-Base-1.5B2026.02 | 32.6 | 1.21 | |
| No CoTBase Model=Qwen2.5-Base-1.5B2026.02 | 31.92 | 2.03 |