LLM-as-a-judge alignment on Clinician Validation Obesity cohort (R1)
85.73-Way AgreementQwen3-32B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Qwen3-32B2026.05 | 85.7 | 0.64 | |
| MedGemma-27B-text-it2026.05 | 85 | 0.53 | |
| GPT-4.12026.05 | 84.2 | 0.62 | |
| GPT-4.1-mini2026.05 | 78.2 | 0.51 | |
| GPT-52026.05 | 69.2 | 0.4 |