TransMem：将隐藏状态转化为大语言模型的记忆
原标题：TransMem: Transforming Hidden States into Memory for Large Language Models
AI 摘要
TransMem 是一种轻量级推理时参数记忆模块，可将冻结的大语言模型（LLM）的稀疏历史隐藏状态转换为可复用的记忆表示，通过门控网络动态干预当前隐藏状态，无需重复编码先前上下文。实验表明，TransMem 在 LoCoMo、HotpotQA 和 MemoryAgentBench 上均取得显著性能提升，验证了稀疏历史隐藏状态作为长上下文 LLM 智能体有效记忆基质的可行性。
正文节选
Computer Science > Multiagent Systems Title:TransMem: Transforming Hidden States into Memory for Large Language Models View PDF HTML (experimental) Abstract:Large language model (LLM) agents increasingly operate over long interaction histories, where effective reasoning requires identifying and exploiting task-relevant evidence distributed across past observations and actions. However, useful information encoded in previously computed representations is often underutilized during sub