LATIDIA · Investigación

Modelo de recompensa de progreso para el aprendizaje de refuerzo a través de modelos de lenguaje grandes

arXiv: 2610.22056v1Tipo de anuncio: nuevo Resumen: Los algoritmos tradicionales de aprendizaje por refuerzo (RL) enfrentan limitaciones significativas en el manejo de tareas a largo plazo con recompensas escasas. Los avances recientes han aprovechado la gran la

La noticia

arXiv:2610.22056v1 Announce Type: new Abstract: Traditional reinforcement learning (RL) algorithms face significant limitations in handling long-term tasks with sparse rewards. Recent advancements have leveraged large language models (LLMs) to enhance RL by utilizing their world knowledge for task planning and reward generation. However, planning-based approaches often depend on pre-defined skill libraries and fail to optimize low-level control policies, while reward-based methods require extensive human feedback or exhaustive searching due to the complexity of tasks. In this paper, we propose the Progress Reward Model for RL (PRM4RL), a novel framework that integrates task planning and dense reward to enhance RL. For high-level planning, a complex task is decomposed into a series