Open-domain Question Answering on COMPMIX first 100 samples (dev)
92Platinum ScoreSPAGHETTI
Evaluation Results
| Method | Links | |
|---|---|---|
| SPAGHETTIBackbone=GPT-42024.06 | 92 | |
| SPAGHETTIBackbone=GPT-3.52024.06 | 84 | |
| GPT-4Zero-shot=true, Engine=GPT-42024.06 | 81 | |
| SPAGHETTIBackbone=LLaMA-7B2024.06 | 77 | |
| GPT-3.5 (turbo-instruct)Zero-shot=true, Engine=turbo-instruct2024.06 | 74 |