DELTAMEM: Incremental Experience Memory for LLM Agents via Residual Trees

In three linesDeltaMem organizes LLM agent experience memory into two residual trees: one stores goal-conditioned tasks as reusable skills, another stores scene-level environment knowledge. Each tree uses root nodes for generalized base experiences and delta nodes for variations, eliminating redundancy. An autonomous consolidation mechanism distills high-frequency paths into new root nodes.

Summary generated by Claude — human-verified