Generalized Preference Optimization: A Unified Approach to Offline Alignment

Documenti analoghi

• Human Alignment of Large Language Models through Online Preference Optimisation
  di: Calandriello, Daniele, et al.
  Pubblicazione: (2024)
• VA-learning as a more efficient alternative to Q-learning
  di: Tang, Yunhao, et al.
  Pubblicazione: (2023)
• Offline Regularised Reinforcement Learning for Large Language Models Alignment
  di: Richemond, Pierre Harvey, et al.
  Pubblicazione: (2024)
• Understanding the performance gap between online and offline alignment algorithms
  di: Tang, Yunhao, et al.
  Pubblicazione: (2024)
• Off-policy Distributional Q($λ$): Distributional RL without Importance Sampling
  di: Tang, Yunhao, et al.
  Pubblicazione: (2024)