发表机构
Jilin University; Xiamen University（吉林大学; 厦门大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对长期记忆引发智能体谄媚行为的问题，提出MemAdapter框架，通过反事实归纳、情境感知反思和基于证据的推理自适应整合记忆，在三个基准上提升记忆可靠性。
AI 中文摘要
长期记忆使基于LLM的智能体能够在任务和会话之间保留并重用信息，支持个性化和长时程交互。然而，持久记忆也可能引发谄媚行为，导致智能体过度迎合用户的历史信念，即使这些信念不准确、过时或与客观证据不一致。现有的缓解方法假设记忆引发的谄媚源于有偏或错误的记忆，并试图通过在记忆管道的不同阶段过滤此类记忆来降低风险。然而，在现实世界中，客观且正确的记忆仍可能引发谄媚行为，且同一记忆在不同情境下可能产生不同的影响。为此，我们提出了MemAdapter，一种新颖的框架，能够自适应地整合检索到的记忆，以支持客观且可靠的推理。具体而言，MemAdapter包含三个组件：（i）反事实归纳，利用反事实推理揭示检索记忆的潜在风险；（ii）情境感知反思，通过自我反思根据当前任务校准每条检索记忆的推理影响；（iii）基于证据的推理，在保留记忆合法影响的同时，将最终回答建立在适当的证据之上。在三个基准上的大量实验表明，MemAdapter在不同场景下持续提升了记忆可靠性。我们的代码可在以下网址获取：此https URL。
英文摘要
Long-term memory enables LLM-based agents to retain and reuse information across tasks and sessions, supporting personalization and long-horizon interactions. However, persistent memories can also induce sycophancy, causing agents to over-align with users' historical beliefs even when they are inaccurate, outdated, or inconsistent with objective evidence. Existing mitigation methods assume that memory-induced sycophancy originates from biased or incorrect memories and attempt to reduce this risk by filtering such memories at different stages of the memory pipeline. However, in the real world, objective and correct memories can still induce sycophancy, and the same memory can warrant different influence across different contexts. To this end, we propose MemAdapter, a novel framework that adaptively integrates retrieved memories to support objective and reliable reasoning. Specifically, MemAdapter consists of three components: (i) Counterfactual Induction, which leverages counterfactual reasoning to uncover the potential risk of retrieved memories; (ii) Context-Aware Reflection, which calibrates the inferential influence of each retrieved memory in light of the current task via self-reflection; and (iii) Evidence-Based Reasoning, which grounds the final response in appropriate evidence while preserving the legitimate influence of memory. Extensive experiments on three benchmarks demonstrate that MemAdapter consistently improves memory reliability across diverse scenarios. Our code is available at https://github.com/DEEP-JLU/MemAdapter.
