SF-AMS: Strategic Forgetting for Structured Memory in LLM Agent

Researchers propose SF-AMS, a framework for strategic forgetting in LLM agent memory systems that replaces static retrieval with a utility-driven survival mechanism, achieving consistent gains over baselines including LightMem, MemO, and A-Mem. On LoCoMo and LongMemEval-s, SF-AMS improves multi-hop reasoning under Qwen2.5-7B by +9.65 F1, temporal reasoning under GPT-4o-mini by +6.91 F1, and open-domain tasks by +6.53 F1, demonstrating strong cross-backbone generalization.

arXiv:2607.22562v1 Announce Type: new Abstract: Managing long-context dependencies remains a primary bottleneck in LLM agents, as redundant and irrelevant information can degrade multi-step reasoning. Strategic Forgetting for Agent Memory Systems SF-AMS is proposed as a framework for maintaining compact high-utility memory by modeling the long-term importance of memory units. SF-AMS replaces static retrieval and heuristic decay with a utility-driven survival mechanism that updates memory importance from usage redundancy and temporal signals, inducing a hierarchical memory structure that prioritizes stable entity-consistent information while filtering noise. On top of this, Composite Importance Scoring integrates semantic and entity level signals to improve retrieval robustness. Experiments on LoCoMo and LongMemEval-s show consistent gains over strong state of the art baselines including LightMem MemO and A-Mem. The largest improvement appears in multi-hop reasoning under Qwen2.5-7B where SF-AMS achieves plus 9.65 F1 over the strongest baseline followed by temporal reasoning under GPT-4o-mini plus 6.91 F1 and open-domain tasks plus 6.53 F1 demonstrating strong cross backbone generalization. These results show that modeling memory importance as a dynamic utility signal is critical for reliable long-context reasoning.