Question Answering on MultiFieldQA-zh
35.25F1 ScoreDISRetrieval
Evaluation Results
| Method | Links | |
|---|---|---|
| DISRetrievalReader Model=GPT-4.1-mini, Budget=4002025.05 | 35.25 | |
| BisectionReader Model=GPT-4.1-mini, Budget=4002025.05 | 34.8 | |
| flatten-sentenceReader Model=GPT-4.1-mini, Budget=4002025.05 | 34.41 | |
| RAPTORReader Model=GPT-4.1-mini, Budget=4002025.05 | 34.32 | |
| flatten-chunkReader Model=GPT-4.1-mini, Budget=4002025.05 | 33.46 | |
| DISRetrievalReader Model=GPT-4.1-mini, Budget=3002025.05 | 32.97 | |
| BisectionReader Model=GPT-4.1-mini, Budget=3002025.05 | 32.31 | |
| flatten-sentenceReader Model=GPT-4.1-mini, Budget=3002025.05 | 31.95 | |
| RAPTORReader Model=GPT-4.1-mini, Budget=3002025.05 | 31.7 | |
| flatten-chunkReader Model=GPT-4.1-mini, Budget=3002025.05 | 31.25 | |
| DISRetrievalReader Model=GPT-4.1-mini, Budget=2002025.05 | 29.6 | |
| DISRetrievalReader Model=Deepseek-v3, Budget=4002025.05 | 29.54 | |
| flatten-chunkReader Model=GPT-4.1-mini, Budget=2002025.05 | 28.77 | |
| DISRetrievalReader Model=Deepseek-v3, Budget=3002025.05 | 28.61 | |
| flatten-sentenceReader Model=GPT-4.1-mini, Budget=2002025.05 | 28.59 | |
| BisectionReader Model=Deepseek-v3, Budget=4002025.05 | 28.24 | |
| BisectionReader Model=GPT-4.1-mini, Budget=2002025.05 | 28.23 | |
| flatten-sentenceReader Model=Deepseek-v3, Budget=4002025.05 | 27.06 | |
| BisectionReader Model=Deepseek-v3, Budget=3002025.05 | 27.05 | |
| flatten-sentenceReader Model=Deepseek-v3, Budget=3002025.05 | 27.04 | |
| RAPTORReader Model=Deepseek-v3, Budget=4002025.05 | 27.01 | |
| flatten-chunkReader Model=Deepseek-v3, Budget=3002025.05 | 26.91 | |
| DISRetrievalReader Model=Deepseek-v3, Budget=2002025.05 | 26.76 | |
| BisectionReader Model=Deepseek-v3, Budget=2002025.05 | 26.71 | |
| flatten-chunkReader Model=Deepseek-v3, Budget=4002025.05 | 26.7 | |
| RAPTORReader Model=GPT-4.1-mini, Budget=2002025.05 | 26.61 | |
| flatten-sentenceReader Model=Deepseek-v3, Budget=2002025.05 | 26.15 | |
| RAPTORReader Model=Deepseek-v3, Budget=3002025.05 | 25.96 | |
| RAPTORReader Model=Deepseek-v3, Budget=2002025.05 | 23.11 | |
| flatten-chunkReader Model=Deepseek-v3, Budget=2002025.05 | 22.69 |