Scientific Claim Verification on NealSmith abstract-level (test)
69.03SUP F1 ScoreDeepSeek-R1
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| DeepSeek-R1Evaluation Protocol=Zero-shot2026.06 | 69.03 | 16.97 | 23.95 | 36.65 | |
| DeepSeek-V3Evaluation Protocol=Zero-shot2026.06 | 60.99 | 13 | 22.93 | 33.3 |