MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Use

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Mengru Wang, Haozhe Luo, Zhenqian Xu, Zhixiang Cui, Haoming Xu, Qu Yang, Jizhan Fang, Junfeng Fang, Ningyu Zhang
• Venue: arXiv: cs.AI
• Published: 2026-08-20
• Collected: 2026-08-21
• URL: http://arxiv.org/abs/2608.20202v1
• PDF: https://arxiv.org/pdf/2608.20202v1
• DOI: UNKNOWN
• Registry group: academic_apis_and_preprints
• Evidence role: canonical_academic_index
• Language: en
• Country / region: UNKNOWN

Plain-Language Brief

Researchers have introduced a new benchmark called MemTrapBench to test the performance of large language models (LLMs) in terms of cognitive traps caused by their memory components. The study identifies two types of cognitive traps: Reasoning Fixation and Belief Distortion, which can affect the model's reasoning and current task performance even when memories are correctly recorded and semantically relevant. The researchers propose a method called AdaptiveMem to mitigate these cognitive traps, improving or preserving the performance on standard memory benchmarks.

What the Document Says

• Memory is a key component in large language models.
• Existing memory benchmarks focus on information extraction, storage, and retrieval but overlook how memories affect current task performance and reasoning.
• The study identifies two types of cognitive traps: Reasoning Fixation and Belief Distortion.
• MemTrapBench is introduced to evaluate these failure modes across two model families and five memory frameworks.
• All evaluated memory strategies underperform the no-memory setting, with drops exceeding 10%.
• AdaptiveMem is proposed as a method to mitigate cognitive traps on MemTrapBench while maintaining or improving performance on standard memory benchmarks.

Key Concepts Mentioned

• large language models
• memory benchmarks
• cognitive traps
• Reasoning Fixation
• Belief Distortion
• MemTrapBench
• AdaptiveMem

Claims Made by the Document

• Existing memory benchmarks overlook how memories affect current task performance and reasoning.
• All evaluated memory strategies underperform the no-memory setting on MemTrapBench.
• AdaptiveMem mitigates cognitive traps on MemTrapBench.

Evidence Used

• The text states that existing memory benchmarks mainly evaluate information extraction, storage, and retrieval but largely overlook how retrieved memories reshape model reasoning and affect performance.
• The study shows that all evaluated memory strategies underperform the no-memory setting on MemTrapBench, with drops exceeding 10%.
• AdaptiveMem is presented as a method to mitigate cognitive traps on MemTrapBench.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

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.