paper-with-me

Papers

FlipLLM: Efficient Bit-Flip Attacks on Multimodal LLMs using Reinforcement Learning

2025-12-10 · Khurram Khalil, Khaza Anuarul Hoque arxiv

Generative Artificial Intelligence models, such as Large Language Models (LLMs) and Large Vision Models (VLMs), exhibit state-of-the-art performance but remain vulnerable to hardware-based threats, specifically bit-flip attacks (BFAs). Existing BFA discovery methods lack generalizability and struggle to scale, often failing to analyze the vast parameter space and complex interdependencies of modern foundation models in a reasonable time. This paper proposes FlipLLM, a reinforcement learning (RL) architecture-agnostic framework that formulates BFA discovery as a sequential decision-making problem. FlipLLM combines sensitivity-guided layer pruning with Q-learning to efficiently identify minimal, high-impact bit sets that can induce catastrophic failure. We demonstrate the effectiveness and generalizability of FlipLLM by applying it to a diverse set of models, including prominent text-only LLMs (GPT-2 Large, LLaMA 3.1 8B, and DeepSeek-V2 7B), VLMs such as LLaVA 1.6, and datasets, such as MMLU, MMLU-Pro, VQAv2, and TextVQA. Our results show that FlipLLM can identify critical bits that are vulnerable to BFAs up to 2.5x faster than SOTA methods. We demonstrate that flipping the FlipLLM-identified bits plummets the accuracy of LLaMA 3.1 8B from 69.9% to ~0.2%, and for LLaVA's VQA score from 78% to almost 0%, by flipping as few as 5 and 7 bits, respectively. Further analysis reveals that applying standard hardware protection mechanisms, such as ECC SECDED, to the FlipLLM-identified bit locations completely mitigates the BFA impact, demonstrating the practical value of our framework in guiding hardware-level defenses. FlipLLM offers the first scalable and adaptive methodology for exploring the BFA vulnerability of both language and multimodal foundation models, paving the way for comprehensive hardware-security evaluation.

📄 PDF Abstract BibTeX arXiv:2512.09872

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

BitHydra: Towards Bit-flip Inference Cost Attack against Large Language Models

2025-05-22 · Xiaobei Yan, Yiming Li, Zhaoxin Fan, Han Qiu 외

Large language models (LLMs) have shown impressive capabilities across a wide range of applications, but their ever-increasing size and resource demands make them vulnerable to inference cost attacks, where attackers ind…

Beyond Jailbreaks: Revealing Stealthier and Broader LLM Security Risks Stemming from Alignment Failures

2025-06-09 · Yukai Zhou, Sibei Yang, Wenjie Wang

Large language models (LLMs) are increasingly deployed in real-world applications, raising concerns about their security. While jailbreak attacks highlight failures under overtly harmful queries, they overlook a critical…

FlipGuard: Defending Large Language Models Against Quantization-Conditioned Backdoor Attacks

2026-06-27 · Aoying Zheng, Anqi Du, Zizhuang Deng, Yuxuan Chen arxiv

Model quantization is essential for the efficient deployment of Large Language Models (LLMs), but introduces a critical vulnerability: Quantization-Conditioned Backdoor (QCB) attacks. In these attacks, malicious behavior…

Code Generation

QFlip: An Adaptive Reinforcement Learning Strategy for the FlipIt Security Game

2019-06-27 · Lisa Oakley, Alina Oprea

A rise in Advanced Persistent Threats (APTs) has introduced a need for robustness against long-running, stealthy attacks which circumvent existing cryptographic security guarantees. FlipIt is a security game that models …

OpenAI GymQ-Learningreinforcement-learningReinforcement Learning+1

Same Evidence, Different Answer: Auditing Order Sensitivity in Multimodal Large Language Models

2026-06-24 · Akshay Paruchuri, Sanmi Koyejo, Ehsan Adeli arxiv

Standard benchmarks for multimodal large language models (MLLMs) score each item on one canonical ordering and miss whether order-irrelevant shuffling changes the answer, a baseline reliability property called for by eme…

Visual Reasoning