Scientific Claim Verification on SciFact
40.5Accuracymistral-large
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| mistral-largeModel Family=Mistral, Size=Large2026.02 | 40.5 | 19.2 | 22.1 | 50.6 | 54.9 | |
| deepseek-chatModel Family=DeepSeek2026.02 | 40 | 19.1 | 22.4 | 50.6 | 56.1 | |
| llama-3.3-70bModel Family=Llama 3, Size=70B2026.02 | 40 | 19 | 22.5 | 50.6 | 55.6 | |
| flan-t5-largeModel Family=FLAN-T5, Size=Large2026.02 | 39.5 | 18.9 | 22.8 | 49.6 | 55.1 | |
| gpt-4o-miniModel Family=GPT-4o, Size=Mini2026.02 | 39.5 | 18.9 | 22.3 | 50.6 | 56.1 | |
| gpt-5.2-chatModel Family=GPT-52026.02 | 39.5 | 18.9 | 22.6 | 50.6 | 55.9 |