SOURCE-LINKED INTELLIGENCE

CHIME: Credit-Aware Hierarchical Memory Evolution for Long-Horizon Agentic Planning

arXiv · AI, language, vision and robotics · article · Sep 2, 2026 · UTC

Planning is a central capability that enables agents to decompose complex long-horizon tasks into manageable steps. Test-time search and training-based methods improve planning but incur high inference costs or require expensive training data. Self-evolving memory instead accumulates reusable experience from agent interaction outcomes into an external memory bank, so planning capability keeps improving at inference time without parameter updates. However, existing self-evolving memory methods share an inherent credit assignment problem: they rely on final task outcomes as feedback, but such ou

Read original source ↗ Open in workspace

recordType

paper

region

Global