Clinical Counseling Dialogue Evaluation on Session Rating Scale Resistant
3.48Helpful RatingSTEPPERSFT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| STEPPERSFTPlanning=Enabled, Supervised Fine-Tuning=Enabled2026.04 | 3.48 | 1.95 | |
| STEPPERSFT + PREF.Planning=Enabled, Supervised Fine-Tuning=Enabled, Preference Tuning=Enabled2026.04 | 3.48 | 1.95 | |
| CAMEL2026.04 | 3.16 | 2.14 | |
| LLAMA-PSYCH8K2026.04 | 3.12 | 2.15 | |
| STEPPERSFT_NOPLANPlanning=Disabled, Supervised Fine-Tuning=Enabled2026.04 | 3.05 | 2.2 | |
| GPT-4O2026.04 | 3.02 | 2.26 | |
| GEMINI-2.0-FLASH2026.04 | 2.81 | 2.55 | |
| SMILECHAT2026.04 | 2.67 | 2.83 | |
| CBT-LLM2026.04 | 2.32 | 3.15 |