LATIDIA · Robótica
LLM-Flax: Reglas generadas por LLM y orientación sobre objetos para la planificación de tareas neurosimbólicas
arXiv:2604.26569v2 Tipo de anuncio: sustitución Resumen: Adaptar un planificador de tareas neurosimbólico a un nuevo dominio puede requerir reglas simbólicas creadas manualmente e instancias de entrenamiento resueltas para un evaluador de objetos aprendido. Este artículo pr
WhatsApp ↗Telegram ↗
La noticia
arXiv:2604.26569v2 Announce Type: replace Abstract: Adapting a neuro-symbolic task planner to a new domain can require hand-authored symbolic rules and solved training instances for a learned object scorer. This paper presents LLM-Flax, which applies large language models (LLMs) to offline rule generation, failure recovery, and object scoring within the Flax planning pipeline. Generated rules undergo validation and corrective feedback before reuse, while runtime guidance operates within a shared planning budget. Experiments on MazeNamo and two additional domains show that the evaluated rule configuration improves planning success over manually authored rules while retaining the trained object scorer. Across eight MazeNamo settings, the mean success rate increases by approximately three percentage points.