---
title: "MemTrapBench：评估大语言模型记忆中的认知陷阱"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.20202v1"
permalink: "https://feed.astralor.com/items/itm_bb88ef095c9b7c00e38262d8"
published: "2026-08-20T16:00:17.000Z"
score: 76
channel: "ai"
tags: ["AdaptiveMem","LLM","MemTrapBench","认知陷阱","记忆评测"]
---

# MemTrapBench：评估大语言模型记忆中的认知陷阱

## AI 点评

> 这篇论文最值得看的是它揭示了“忠实记忆也能坑模型”的盲区，提醒做记忆系统的人别只盯着存取精度。对于正在构建长期 Agent 记忆的工程师和研究者来说，MemTrapBench 和 AdaptiveMem 都是绕不开的参考。

## 速览

现有大语言模型记忆评测主要关注信息提取、存储与检索的准确性，却忽视了检索到的记忆会重塑模型推理并影响当前任务表现。MemTrapBench 聚焦两类记忆诱发的认知陷阱：推理固化和信念扭曲。实验显示，现有记忆策略在 MemTrapBench 上甚至不如无记忆设置，最强方法性能下降超 10%。作者提出 AdaptiveMem，一种简单的推理期干预方法，可有效缓解认知陷阱，同时保持或提升标准记忆评测上的表现。

## 关键点

- MemTrapBench 覆盖两类认知陷阱：推理固化和信念扭曲，用于测试记忆对当前推理的负面影响。
- 实验证明现有记忆框架在 MemTrapBench 上均不如无记忆基线，最强方法也下降超 10%。
- AdaptiveMem 通过推理期指令引导模型避开记忆陷阱，在缓解问题的同时不损害标准记忆任务。

## 全文

Memory has become a key component of large language models, enabling them to retain information and learn from long-term interactions. However, existing memory benchmarks mainly evaluate whether information is correctly extracted, stored, and retrieved, while largely overlooking how retrieved memories reshape model reasoning and affect performance on the current task. We identify memory-induced cognitive traps: even faithfully recorded and semantically relevant memories can distort model reasoning or beliefs and degrade current task performance. To systematically evaluate these failure modes, we introduce MemTrapBench, which covers two forms of cognitive traps: Reasoning Fixation and Belief Distortion. Experiments across two model families and five representative memory frameworks show that MemTrapBench is challenging: all evaluated memory strategies underperform the no-memory setting, with even the strongest methods suffering drops of more than 10%. To mitigate these cognitive traps, we propose AdaptiveMem, a simple yet effective inference-time method that instructs LLMs to avoid memory traps. AdaptiveMem mitigates cognitive traps on MemTrapBench while preserving or improving performance on standard memory benchmarks across diverse memory frameworks.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_bb88ef095c9b7c00e38262d8