AI RESEARCH
PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement
arXiv CS.AI
•
ArXi:2605.11225v1 Announce Type: new Large language model (LLM)-based agents frequently generate seemingly coherent plans that fail upon execution due to infeasible actions, constraint violations, and compounding errors over extended horizons. PIVOT (Plan-Inspect-eVOlve Trajectories) addresses this plan-execution misalignment through a self-supervised framework that treats trajectories as optimizable objects iteratively refined via environment interaction.