LATIDIA · Investigación

Compromiso de planificación de LLM bajo observabilidad parcial

arXiv: 2610.21108v1Announce Type: new Abstract: Los agentes de planificación basados en LLM se comprometen con planes de diferentes longitudes antes de interactuar con un entorno parcialmente observable. Algunos se comprometen agresivamente con un plan a largo plazo basado en

La noticia

arXiv:2610.21108v1 Announce Type: new Abstract: LLM-based planning agents commit to plans of different lengths before interacting with a partially observable environment. Some aggressively commit to a long plan based on assumptions about unobserved states, while others conservatively plan only over what they have observed. Both behaviors emerge implicitly in pre-trained models, whether the LLM outputs a plan directly or generates a PDDL formulation for a classical solver. Aggressive commitment risks costly failures, while conservative commitment risks excessive LLM calls. This choice pits safety against efficiency, a trade-off that has been neither characterized systematically nor made controllable. We investigate aggressive and conservative LLM planners and formalizers on two partially observable domains,