加权记忆树：为长视野大语言模型智能体记住重要信息
Weighted Memory Tree: Remembering What Matters for Long-Horizon LLM Agents
浏览论文内容
中文总结 AI 辅助
提出加权记忆树（WMT）分层记忆系统，在GAIA-Text数据集上使LLM智能体准确率平均提升9.97个百分点、令牌用量减32.8%，可抑制低效用内容与不可靠信息传播。
中文摘要 AI 辅助
大语言模型（LLM）智能体已展现出解决多步骤任务的能力，这些任务需要规划、工具使用和外部信息访问，但不断增长的执行历史会增加推理成本，并使推理面临过时、无关或误导性信息的影响，可能会降低推理质量。现有的记忆方法会组织或压缩执行历史，但在决定哪些记忆保持活跃方面的机制有限。我们提出了加权记忆树（WMT），这是一种分层记忆系统，将执行过程组织为任务、子任务和动作，同时为每个记忆分配动态保留分数。基于事件的更新和基于选择的衰减会修改这些分数，使WMT能够保留有用信息、折叠完成的轨迹、抑制低效用内容，并保留对折叠上下文的访问权限。我们使用Qwen3-8B、Gemma 4 E4B和Llama-3.1-8B在GAIA-Text数据集上对WMT进行评估，包含 ablation 实验和记忆投毒实验。与线性记忆相比，WMT的准确率平均提高了9.97个百分点，同时提示词令牌使用量减少了32.8%。记忆投毒实验表明，WMT限制了不可靠信息的持久性和传播。我们的结果表明，有效的长视野智能体记忆更多取决于决定哪些信息应保持活跃，而非存储更多信息。
英文摘要
Large language model (LLM) agents have demonstrated the ability to solve multi-step tasks requiring planning, tool use, and external information access, yet growing execution histories increase inference cost and expose reasoning to outdated, irrelevant, or misleading information, potentially degrading reasoning quality. Existing memory approaches organize or compress execution histories but provide limited mechanisms for deciding which memories remain active. We introduce the, a hierarchical memory system that organizes execution into tasks, subtasks, and actions while assigning each memory a dynamic retention score. Event-based updates and selection-based decay revise these scores, allowing WMT to preserve useful information, fold completed trajectories, suppress low-utility content, and retain access to folded context. We evaluate WMT on GAIA-Text using Qwen3-8B, Gemma 4 E4B, and Llama-3.1-8B, with ablations and memory-poisoning experiments. Relative to linear memory, WMT improves accuracy by an average of 9.97 percentage points while reducing prompt-token usage by 32.8%. Memory-poisoning experiments show that WMT limits the persistence and propagation of unreliable information. Our results suggest that effective long-horizon agent memory depends less on storing more information than on deciding which information should remain active.
发表机构
- Rose-Hulman Institute of Technology（罗斯-霍曼理工学院）
- Georgia Institute of Technology（佐治亚理工学院）
- Georgia Tech Research Institute（佐治亚理工学院研究院）
机构由 AI 辅助整理，请以论文原文为准。
