arXiv：cs.CL· Taiqiang Wu, Yuxin Cheng, Chenchen Ding, Runming Yang, Xincheng Feng, Wenyong Zhou, Zhengwu Liu, Ngai Wong·· 4 小时前AI 评分31
忆阻器上的 LLM 还能信任吗？非理想性下的推理能力探究
Can We Trust LLMs on Memristors? Diving into Reasoning Ability under Non-Ideality
AI 导读
研究系统考察了忆阻器模拟存内计算架构的非理想性对 LLM 推理能力的影响，发现推理性能显著下降且在不同 benchmark 上差异明显。
正文
Abstract:Memristor-based analog compute-in-memory (CIM) architectures provide a promising substrate for the efficient deployment of Large Language Models (LLMs), owing to superior energy efficiency and computational density. However, these architectures suffer from precision issues caused by intrinsic non-idealities of memristors. In this paper, we first conduct a comprehensive investigation into the impact of such typical non-idealities on LLM reasoning. Empirical results indicate that reasoning capability decreases significantly but varies for distinct benchmarks. Subsequently, we systematically appraise three training-free strategies, including thinking mode, in-context learning, and module redundancy. We thus summarize valuable guidelines, i.e., shallow layer redundancy is particularly effective for improving robustness, thinking mode performs better under low noise levels but degrades at higher noise, and in-context learning reduces output length with a slight performance trade-off. Our findings offer new insights into LLM reasoning under non-ideality and practical strategies to improve robustness.
| Comments: | 7 figures, 3 tables |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2603.13725 [cs.CL] |
| (or arXiv:2603.13725v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2603.13725 arXiv-issued DOI via DataCite |
Submission history
From: Taiqiang Wu [view email]
[v1]
Sat, 14 Mar 2026 03:02:15 UTC (456 KB)
[v2]
Fri, 2 Oct 2026 11:19:36 UTC (421 KB)
来源：arXiv：cs.CL · arxiv.org