Question Answering on TimelineQA (test)
36.9Hit@1RAG
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RAGModel=LLaMA3.3 (70B)2025.05 | 36.9 | 41 | |
| REQAPModel=GPT4o (>100B)2025.05 | 35.7 | 39 | |
| RAGModel=GPT4o (>100B)2025.05 | 32.1 | 36 | |
| REQAPModel=SFT (1B)2025.05 | 31.3 | 35 | |
| REQAPModel=LLaMA3.3 (70B)2025.05 | 28.6 | 31 | |
| CODEGENModel=GPT4o (>100B)2025.05 | 26.2 | 33 | |
| RAGModel=SFT (1B)2025.05 | 21.4 | 22 | |
| CODEGENModel=LLaMA3.3 (70B)2025.05 | 18.7 | 22 | |
| CODEGENModel=SFT (1B)2025.05 | 13.5 | 17 |