MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Use

Memory has become a key component of large language models, enabling them to retain information and learn from longterm 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 nomemory 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 1.

Introduction. Recent memory frameworks have made substantial progress (Huang et al. 2026; Zhang et al. 2025; Yu et al. 2026; Zhang et al. 2026c,b; Tan et al. 2026; Hu et al. 2025) in extending the effective context available to large language models (LLMs) and agents (Du et al. 2025; Wu et al. 2025b; Hu et al. 2025; Chen et al. 2026; Yan et al. 2026). Most existing studies focus on constructing and maintaining external memory from long interaction histories, then retrieving relevant information to support subsequent queries (Behrouz, Zhong, and Mirrokni 2025; Xu et al. 2026a; Chhikara et al. 2025; Tan et al. 2025b; Shen et al. 2026; Uddin et al. 2026; Wang et al. 2025a; Tavakoli et al. 2025). Accordingly, existing benchmarks primarily assess memory extraction, storage, updating, and retrieval (Tan et al. 2025a; Wu et al. 2025a; Hu, Wang, and McAuley 2025). Yet memory is not always what we need, as it may impair rather than enhance model capabilities. Prior

Discussion / Conclusion. We introduce MemTrapBench, a benchmark for evaluating memory-induced cognitive traps in LLMs. Experiments across multiple models and memory strategies show that MemTrapBench remains challenging for existing memory frameworks. We further propose AdaptiveMem, a simple yet effective approach for more reliable memory use.