AI RESEARCH
HiMAC: Hierarchical Macro-Micro Learning for Long-Horizon LLM Agents
arXiv CS.LG
•
ArXi:2603.00977v2 Announce Type: replace-cross Large language model (LLM) agents have recently nstrated strong capabilities in interactive decision-making, yet they remain fundamentally limited in long-horizon tasks that require structured planning and reliable execution. Existing approaches predominantly rely on flat autoregressive policies, where high-level reasoning and low-level actions are generated within a single token sequence, leading to inefficient exploration and severe error propagation over extended trajectories.