让大模型主动遗忘无用信息，提升长程推理能力
SF-AMS: Strategic Forgetting for Structured Memory in LLM Agent
- 用动态实用度机制替代固定记忆策略，自动筛选重要信息
- 在多跳推理中比最强基线高9.65 F1，跨模型通用性强
- 适合需要长期记忆的复杂任务场景，如对话、问答系统
长上下文依赖管理仍是大模型智能体的主要瓶颈，冗余和无关信息会削弱多步推理能力。本文提出策略性遗忘框架SF-AMS，通过建模记忆单元的长期重要性，实现紧凑且高价值的记忆保持。该框架将静态检索与启发式衰减替换为基于实用度的生存机制，结合使用冗余与时间信号动态更新记忆重要性，形成分层记忆结构，优先保留稳定一致的实体信息并过滤噪声。此外，复合重要性评分融合语义与实体级信号，增强检索鲁棒性。在LoCoMo和LongMemEval-s上的实验表明，SF-AMS持续优于包括LightMem、MemO和A-Mem在内的强基线模型。其中，在Qwen2.5-7B上的多跳推理任务中，性能提升达9.65 F1；GPT-4o-mini在时间推理任务上提升6.91 F1；开放域任务提升6.53 F1，体现出色的跨模型泛化能力。结果表明，将记忆重要性建模为动态实用度信号，对可靠长上下文推理至关重要。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。