Question Answering on HQA (in-domain)
39.6EMOriginal Docs
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Original DocsType=N/A, Retrieval depth=Top-20, Reader model=GPT-4o2024.12 | 39.6 | 51.8 | 2,940.5 | 100 | |
| EXITType=Ext., Retrieval depth=Top-20, Reader model=GPT-4o2024.12 | 39.4 | 50.1 | 453.6 | 15.4 | |
| LongLLMLinguaType=Ext., Retrieval depth=Top-20, Reader model=GPT-4o2024.12 | 38.8 | 49.4 | 549.5 | 18.7 | |
| EXITType=Ext., Retrieval depth=Top-5, Reader model=GPT-4o2024.12 | 38.2 | 50.4 | 191.2 | 26 | |
| CompActType=Abs., Retrieval depth=Top-5, Reader model=GPT-4o2024.12 | 37.4 | 48 | 74.3 | 10.1 | |
| Original DocsType=N/A, Retrieval depth=Top-5, Reader model=GPT-4o2024.12 | 37.2 | 48.6 | 735.3 | 100 | |
| RefinerType=Abs., Retrieval depth=Top-5, Reader model=GPT-4o2024.12 | 35.8 | 47.5 | 71.4 | 9.7 | |
| LongLLMLinguaType=Ext., Retrieval depth=Top-5, Reader model=GPT-4o2024.12 | 34.2 | 45.2 | 223.1 | 30.3 | |
| RECOMP-AbstType=Abs., Retrieval depth=Top-20, Reader model=GPT-4o2024.12 | 33.6 | 44.2 | 62.7 | 2.1 | |
| CompActType=Abs., Retrieval depth=Top-20, Reader model=GPT-4o2024.12 | 33 | 43.7 | 106 | 3.6 | |
| RECOMP-ExtrType=Ext., Retrieval depth=Top-5, Reader model=GPT-4o2024.12 | 32.4 | 41.7 | 87.1 | 11.8 | |
| RefinerType=Abs., Retrieval depth=Top-20, Reader model=GPT-4o2024.12 | 31.8 | 41.5 | 130.6 | 4.4 | |
| RECOMP-ExtrType=Ext., Retrieval depth=Top-20, Reader model=GPT-4o2024.12 | 31.2 | 39.6 | 86.2 | 2.9 | |
| RECOMP-AbstType=Abs., Retrieval depth=Top-5, Reader model=GPT-4o2024.12 | 29.4 | 40.2 | 62.8 | 8.5 |