LM-Fix: Lightweight Bit-Flip Detection and Rapid Recovery Framework for Language Models

This paper presents LM-Fix, a lightweight detection and rapid recovery framework for faults in large language models (LLMs). Existing integrity approaches are often heavy or slow for modern LLMs. LM-Fix runs a short test-vector pass and uses hash-guided checks to detect bit-flip faults, then repairs them locally without a full reload. Across multiple models, it detects over 94% of single-bit flips at TVL=200 and nearly 100% of multi-bit flips with approximately 1% to 7.7% runtime overhead; recovery is more than 100x faster than reloading. These results show a practical, low-overhead solution to keep LLMs reliable in production

DOI

10.1109/ICCD65941.2025.00068

Journal

Proc. IEEE Int. Conf. on Computer Design (ICCD), 2025, pp. 432-440

Comment

Accepted at IEEE ICCD 2025. Code: https://github.com/ata990/lm-fix. Detects over 94 percent single-bit flips (near 100 percent multi-bit) with about 1 to 7.7 percent overhead; recovery is over 100x faster than a full reload. Keywords: LLMs, bit-flip, fault injection, reliability, security, Rowhammer, SDC, Jailbreaking, Attack, Defense, GPU DRAM faults