AI RESEARCH

Causal Intervention-Based Memory Selection for Long-Horizon LLM Agents

arXiv CS.CL

ArXi:2605.17641v1 Announce Type: cross Long-horizon LLM agents rely on persistent memory to interactions across sessions, yet existing memory systems often retrieve context using semantic similarity or broad history inclusion, treating retrieved memories as uniformly useful. This assumption is fragile because memories may be topically related while remaining irrelevant, stale, or misleading.