Question Answering on Sports understanding
68.4AccuracyCoT-decoding
Evaluation Results
| Method | Links | |
|---|---|---|
| CoT-decodingModel=Gemma-7B, Spec Ans=Yes2026.04 | 68.4 | |
| GCoT-decoding + SpanAlignModel=Gemma-7B, Spec Ans=No2026.04 | 67.2 | |
| GCoT-decodingModel=Gemma-7B, Spec Ans=No2026.04 | 65.2 | |
| Self-consistencyModel=Gemma-7B, Spec Ans=Yes2026.04 | 63.9 | |
| GreedyModel=Gemma-7B, Spec Ans=No2026.04 | 61.2 | |
| Temperature samplingModel=Gemma-7B, Spec Ans=No2026.04 | 60.1 | |
| Beam searchModel=Gemma-7B, Spec Ans=No2026.04 | 59.9 | |
| Top-k samplingModel=Gemma-7B, Spec Ans=No2026.04 | 58 | |
| GCoT-decodingModel=Llama-3.1-8B, Spec Ans=No2026.04 | 58 | |
| Self-consistencyModel=Llama-3.1-8B, Spec Ans=Yes2026.04 | 54.6 | |
| Self-consistencyModel=Mistral-7B, Spec Ans=Yes2026.04 | 52.9 | |
| Temperature samplingModel=Llama-3.1-8B, Spec Ans=No2026.04 | 52.4 | |
| GCoT-decoding + SpanAlignModel=Llama-3.1-8B, Spec Ans=No2026.04 | 52 | |
| GCoT-decodingModel=Mistral-7B, Spec Ans=No2026.04 | 52 | |
| Top-k samplingModel=Llama-3.1-8B, Spec Ans=No2026.04 | 51.9 | |
| GreedyModel=Llama-3.1-8B, Spec Ans=No2026.04 | 51.6 | |
| CoT-decodingModel=Llama-3.1-8B, Spec Ans=Yes2026.04 | 51 | |
| Beam searchModel=Llama-3.1-8B, Spec Ans=No2026.04 | 50.7 | |
| CoT-decodingModel=Mistral-7B, Spec Ans=Yes2026.04 | 50.6 | |
| Top-k samplingModel=Mistral-7B, Spec Ans=No2026.04 | 50.3 | |
| GreedyModel=Mistral-7B, Spec Ans=No2026.04 | 49.6 | |
| Temperature samplingModel=Mistral-7B, Spec Ans=No2026.04 | 48.9 | |
| Beam searchModel=Mistral-7B, Spec Ans=No2026.04 | 48.2 | |
| GCoT-decoding + SpanAlignModel=Mistral-7B, Spec Ans=No2026.04 | 48 |