Empathetic Dialogue on Chit Chat
83.3ScoreGemini2.5-Pro-0605
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Gemini2.5-Pro-06052025.07 | 83.3 | 77 | 11 | |
| OpenAI-o3-04162025.07 | 83 | 66 | 9 | |
| GPT-4o-03262025.07 | 80.9 | 74 | 17 | |
| GPT-4.1-04142025.07 | 77.1 | 65 | 18 | |
| Gemini2.5-Flash-Think-05202025.07 | 64.7 | 53 | 27 | |
| RLVER (PPO)Think=Yes2025.07 | 62.1 | 52 | 30 | |
| RLVER (PPO)Think=No2025.07 | 53.4 | 39 | 37 | |
| RLVER (GRPO)Think=Yes2025.07 | 53 | 45 | 42 | |
| RLVER (GRPO)Think=No2025.07 | 49.2 | 34 | 40 | |
| Qwen2.5-7B-Instruct2025.07 | 37.8 | 27 | 58 |