Generative Question Answering on SimpleQA
92HALL ScoreRCSP
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| RCSPBackbone=Llama 3.1-8b2026.05 | 92 | 0.67 | 7.33 | 7.35 | |
| Append-idkBackbone=Llama 3.1-8b2026.05 | 91 | 3.33 | 5.67 | 5.76 | |
| EmoPrompt (avg)Backbone=Llama 3.1-8b2026.05 | 91 | 3.73 | 5.27 | 5.38 | |
| BaseBackbone=Llama 3.1-8b2026.05 | 90.67 | 2.67 | 6.67 | 6.76 | |
| BaseBackbone=Gemma-3-12b-pt2026.05 | 87.2 | 3.8 | 9 | 9.17 | |
| RCSPBackbone=Gemma-3-12b-pt2026.05 | 86.33 | 3.33 | 10.33 | 10.51 | |
| Plan-and-SolveBackbone=Llama 3.1-8b2026.05 | 86.33 | 9 | 4.67 | 4.89 | |
| Append-idkBackbone=Gemma-3-12b-pt2026.05 | 85 | 6.67 | 8.33 | 8.62 | |
| EmoPrompt (avg)Backbone=Gemma-3-12b-pt2026.05 | 82.52 | 9.64 | 7.84 | 8.24 | |
| Plan-and-SolveBackbone=Gemma-3-12b-pt2026.05 | 80.67 | 10.33 | 9 | 9.49 |