Preprint
Aug 2026
Online Monitoring and Corrective Steering of Programming Agents
LivePlan decouples judging from advising: a deterministic, rule-based monitor examines general signals over the trajectory to detect issues without invoking an LLM, and only when an issue is detected does it consult an advisor LLM for a high-level, next-step correction.
Shuyang Liu, Saman Dehghan, Ji Young Kim et al.
· 1 citation