Open-domain Question Answering on COMPMIX (dev)
58.5Exact MatchSPAGHETTI
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| SPAGHETTIBackbone=GPT-3.52024.06 | 58.5 | 67.7 | 76.9 | |
| SPAGHETTIBackbone=GPT-42024.06 | 57.3 | 70.2 | 80.8 | |
| SPAGHETTIBackbone=LLaMA-7B2024.06 | 53.8 | 61.7 | 69.8 | |
| GPT-4Zero-shot=true, Engine=GPT-42024.06 | 53 | 60.9 | 76.7 | |
| GPT-3.5 (turbo-instruct)Zero-shot=true, Engine=turbo-instruct2024.06 | 36.4 | 53.2 | 68 |