arXiv Machine Learning June 30, 2026 By Aoying Zheng, Anqi Du, Zizhuang Deng, Yuxuan Chen

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

Read the original on arXiv Machine Learning →

arXiv:2606. 28962v1 Announce Type: cross Abstract: Model quantization is essential for the efficient deployment of Large Language Models (LLMs), but introduces a critical vulnerability: Quantization-Conditioned Backdoor (QCB) attacks.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv AI

Jun 4

Widening the Gap: Exploiting LLM Quantization via Outlier Injection

arXiv:2605. 15152v2 Announce Type: replace-cross Abstract: LLM quantization has become essential for memory-efficient deployment.

By Xiaohua Zhan, Kazuki Egashira, Robin Staab, Mark Vero, Martin Vechev

llms efficiency

arXiv AI

Jun 4

TamperBench: Systematically Stress-Testing LLM Safety Under Fine-Tuning and Tampering

arXiv:2602. 06911v2 Announce Type: replace-cross Abstract: As increasingly capable open-weight large language models (LLMs) are deployed, improving their tamper resistance against unsafe modifications, whether accidental or intentional, becomes critical to minimize risks.

By Saad Hossain, Tom Tseng, Punya Syon Pandey, Samanvay Vajpayee, Matthew Kowal, Nayeema Nonta, Samuel Simko, Stephen Casper, Zhijing Jin, Kellin Pelrine, Sirisha Rambhatla

llms fine-tuning benchmarks safety

arXiv Machine Learning

Jun 4

Bypassing Prompt Guards in Production with Controlled-Release Prompting

arXiv:2510. 01529v3 Announce Type: replace Abstract: Ball et al.

By Jaiden Fairoze, Sanjam Garg, Keewoo Lee, Mingyuan Wang

llms safety

arXiv AI

Jun 3

Patcher: Post-Hoc Patching of Backdoored Large Language Models

arXiv:2606. 02995v1 Announce Type: cross Abstract: Large language models remain vulnerable to jailbreak backdoor attacks, where adversaries poison safety alignment data to embed hidden triggers that bypass safety mechanisms.

By Anjun Gao, Yueyang Quan, Yufei Xia, Zhuqing Liu, Minghong Fang

llms fine-tuning safety

arXiv Machine Learning

Jun 5

Steering Vectors are an Adversarial Attack Surface

arXiv:2606. 05958v1 Announce Type: new Abstract: Activation steering has become a popular way to control Large Language Model (LLM) behavior without fine-tuning.

By Abzal Aidakhmetov, Donato Crisostomi, Tommaso Mencattini, Adrian Robert Minut, Iacopo Masi, Emanuele Rodol\`a

llms fine-tuning multimodal safety

arXiv AI

Jun 11

Grammar-Constrained Decoding Can Jailbreak LLMs into Generating Malicious Code

arXiv:2606. 11817v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly used for code generation, raising concerns that they may be misused to produce malicious code.

By Yitong Zhang, Shiteng Lu, Jia Li

llms benchmarks safety

Related stories

Widening the Gap: Exploiting LLM Quantization via Outlier Injection

TamperBench: Systematically Stress-Testing LLM Safety Under Fine-Tuning and Tampering

Bypassing Prompt Guards in Production with Controlled-Release Prompting

Patcher: Post-Hoc Patching of Backdoored Large Language Models

Steering Vectors are an Adversarial Attack Surface

Grammar-Constrained Decoding Can Jailbreak LLMs into Generating Malicious Code