LLM Preference Alignment Evaluation on HH Helpful
55Preference (spec vs ctl)Spec Learning
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Spec LearningProposer (P)=Gemma 4 31B, Judge Model=GLM-5.12026.06 | 55 | 58 | 70 |
Loading the SOTA2 catalog…
| Method | Links | |||
|---|---|---|---|---|
| Spec LearningProposer (P)=Gemma 4 31B, Judge Model=GLM-5.12026.06 | 55 | 58 | 70 |