Scientific Reasoning on SciKnowEval Hard
42.4Pass@1STV verifier
Evaluation Results
| Method | Links | |
|---|---|---|
| STV verifiergenerator=Qwen3-8B2026.05 | 42.4 | |
| Qwen3-32Bsize_multiplier=4x2026.05 | 25.1 | |
| Qwen3-235B-A22Bsize_multiplier=30x2026.05 | 23.6 | |
| No verificationgenerator=Qwen3-8B2026.05 | 11.5 | |
| No verifier traininggenerator=Qwen3-8B2026.05 | 11.4 |