AI 中文总结
研究大语言模型智能体记忆问题，提出CAMeR框架，结合关键词门控混合激活与自适应权重动态。通过CAMeR-Bench测试，表明其关键词门控效果好，能有效节省令牌并提高检索精度，证明混合符号-神经门控可实现自适应记忆保留。
AI 中文摘要
在长时间对话中运行的大语言模型智能体积累了大量信息，但现有记忆系统要么不加区分地保留所有信息，要么采用统一的遗忘启发式方法，无法区分相关知识和无关知识。我们提出了CAMeR（上下文激活记忆强化），这是一个记忆保留框架，它将关键词门控混合激活（一种联合符号（词级杰卡德）和子符号（嵌入余弦）门控机制）与自适应权重动态相结合。CAMeR为每个记忆查询对计算一个混合相似性分数；超过阈值的记忆会得到强化，而所有记忆都会经历受控衰减。我们引入了CAMeR-Bench，这是一个有76个记忆、100轮的基准测试，涵盖8个主题集群，具有分级激活频率，旨在测试现有基准测试（LoCoMO、LongMemEval）无法测试的自适应保留。在CAMeR-Bench上，与仅使用嵌入门控相比，CAMeR的关键词门在高频和从未引用的记忆之间实现了1.6倍更大的保留差距（剪刀差距：0.039对0.024），而时间驱动的基线（遗忘、超级本地记忆）在100轮后权重降至接近零。CAMeR的前5名检索与全上下文方法相比节省了83.2%的令牌（累积39k对231k），同时产生的权重信号提高了检索精度。通过8种消融条件，我们确定关键词门控而非可学习衰减是此规模下的主要性能驱动因素。我们的发现表明，混合符号-神经门控为大语言模型智能体中的自适应记忆保留提供了一种简单而有效的机制。
英文摘要
Large language model (LLM) agents operating over extended dialogues accumulate vast amounts of information, yet existing memory systems either retain everything indiscriminately or apply uniform forgetting heuristics that fail to distinguish relevant from irrelevant knowledge. We present CAMeR (Context-Activated Memory Reinforcement), a memory retention framework combining keyword-gated hybrid activation -- a joint symbolic (word-level Jaccard) and sub-symbolic (embedding cosine) gating mechanism -- with adaptive weight dynamics. CAMeR computes a hybrid similarity score for each memory-query pair; memories exceeding a threshold receive reinforcement while all memories undergo controlled decay. We introduce CAMeR-Bench, a 76-memory, 100-round benchmark spanning 8 topic clusters with graded activation frequency, designed to test adaptive retention where existing benchmarks (LoCoMO, LongMemEval) cannot. On CAMeR-Bench, CAMeR's keyword gate achieves a 1.6$\times$ larger retention gap between high-frequency and never-referenced memories compared to embedding-only gating (scissors gap: 0.039 vs. 0.024), while time-driven baselines (Oblivion, SuperLocalMemory) collapse to near-zero weights over 100 rounds. CAMeR's top-5 retrieval saves 83.2\% tokens versus full-context approaches (39k vs. 231k cumulative) while producing weight signals that improve retrieval precision. Through 8 ablation conditions we establish that the keyword gate -- not learnable decay -- is the primary performance driver at this scale. Our findings demonstrate that hybrid symbolic-neural gating provides a simple yet effective mechanism for adaptive memory retention in LLM agents.
