Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learning

Documenti analoghi

• mimic-video: Video-Action Models for Generalizable Robot Control Beyond VLAs
  di: Pai, Jonas, et al.
  Pubblicazione: (2025)
• Unsupervised Zero-Shot Reinforcement Learning via Functional Reward Encodings
  di: Frans, Kevin, et al.
  Pubblicazione: (2024)
• Reward Generation via Large Vision-Language Model in Offline Reinforcement Learning
  di: Lee, Younghwan, et al.
  Pubblicazione: (2025)
• On Zero-Shot Reinforcement Learning
  di: Jeen, Scott
  Pubblicazione: (2025)
• Revisiting the Learning Objectives of Vision-Language Reward Models
  di: Roy, Simon, et al.
  Pubblicazione: (2025)