LLM Evaluation on AlpacaEval 2.0 (Quality and Calibration)
84.3Quality ScoreSEE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SEEJudge=Gemini 3.1 Flash-Lite2026.06 | 84.3 | 69 | |
| Adapted RLCRJudge=Gemini 3.1 Flash-Lite2026.06 | 82.9 | 65.5 | |
| Qwen 3-4BJudge=Gemini 3.1 Flash-Lite2026.06 | 82.8 | 65 | |
| SEEJudge=Claude Sonnet 4.62026.06 | 66 | 70.4 | |
| Adapted RLCRJudge=Claude Sonnet 4.62026.06 | 64.3 | 64.5 | |
| Qwen 3-4BJudge=Claude Sonnet 4.62026.06 | 63.7 | 60.9 |