Question Answering on MuSiQue (Cover Exact Match)
24.3Cover EMPAGER
Evaluation Results
| Method | Links | |
|---|---|---|
| PAGERBackbone=Llama3.1-70B-Instruct2026.01 | 24.3 | |
| PAGERBackbone=Qwen3-32B2026.01 | 23 | |
| Search-o1Backbone=Qwen3-32B2026.01 | 22.9 | |
| Search-o1Backbone=Llama3.1-70B-Instruct2026.01 | 22.9 | |
| RATBackbone=Llama3.1-70B-Instruct2026.01 | 20.8 | |
| DeepNoteBackbone=Llama3.1-70B-Instruct2026.01 | 20.8 | |
| StructRAGBackbone=Llama3.1-70B-Instruct2026.01 | 19.4 | |
| Vanilla RAGBackbone=Llama3.1-70B-Instruct2026.01 | 19.1 | |
| DeepNoteBackbone=Qwen3-32B2026.01 | 17.2 | |
| RATBackbone=Qwen3-32B2026.01 | 16.1 | |
| IRCoTBackbone=Qwen3-32B2026.01 | 14.7 | |
| Iter-RetGenBackbone=Qwen3-32B2026.01 | 14.5 | |
| Vanilla LLMBackbone=Llama3.1-70B-Instruct2026.01 | 14.1 | |
| Vanilla RAGBackbone=Qwen3-32B2026.01 | 13.5 | |
| StructRAGBackbone=Qwen3-32B2026.01 | 12.8 | |
| Iter-RetGenBackbone=Llama3.1-70B-Instruct2026.01 | 11.9 | |
| Vanilla LLMBackbone=Qwen3-32B2026.01 | 8.2 | |
| IRCoTBackbone=Llama3.1-70B-Instruct2026.01 | 6.2 |