Scientific Fact-Checking on CLIMATE-FEVER 2-way Supported Refuted subsets
73.83Balanced AccuracyAtomic+Search
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Atomic+Searchsearch=enabled2026.04 | 73.83 | 62.14 | 70.04 | |
| GPT-5 Mini + Searchsearch=enabled2026.04 | 71.2 | 58.5 | 67 | |
| RARR2026.04 | 70.4 | 57.8 | 66.3 | |
| GPT-o12026.04 | 69.7 | 56.1 | 65.05 | |
| MiniCheck2026.04 | 69.1 | 55 | 64 | |
| GPT-5 Mini2026.04 | 67.9 | 54.2 | 63.1 | |
| Qwen 32B MAD2026.04 | 67.3 | 53.6 | 62.6 |