Scientific Fact-Checking on PubMedFact1k 3-way
73.7Macro F1Atomic+Search
Evaluation Results
| Method | Links | |
|---|---|---|
| Atomic+Searchsearch=enabled2026.04 | 73.7 | |
| GPT-5 Mini + Searchsearch=enabled2026.04 | 72.5 | |
| RARR2026.04 | 72.3 | |
| GPT-o12026.04 | 71.2 | |
| GPT-5 Mini2026.04 | 68.5 | |
| Qwen 32B MAD2026.04 | 61.8 |