Noise-Aware Direct Preference Optimization for RLAIF
Reinforcement Learning from Human Feedback (RLHF) produces powerful instruction-following models but relies on a preference-labeling process that is both costly and slow. An effective alternative, Reinforcement Learning from AI Feedback (RLAIF), uses large language models as teachers for relabeling;...
Gardado en:
| Principais autores: | , , , |
|---|---|
| Formato: | Artigo |
| Idioma: | Inglês |
| Publicado: |
MDPI AG
2025-09-01
|
| Series: | Applied Sciences |
| Assuntos: | |
| Acceso en liña: | https://www.mdpi.com/2076-3417/15/19/10328 |
| Tags: |
Sen Etiquetas, Sexa o primeiro en etiquetar este rexistro!
|