English
Related papers

Related papers: Beyond Attack Success Rate: Temporal Logit Observa…

200 papers

Large Language Models have shown impressive generative capabilities across diverse tasks, but their safety remains a critical concern. Existing post-training alignment methods, such as SFT and RLHF, reduce harmful outputs yet leave LLMs…

Cryptography and Security · Computer Science 2025-10-21 Zhengyue Zhao , Yingzi Ma , Somesh Jha , Marco Pavone , Patrick McDaniel , Chaowei Xiao

Jailbreak attacks expose vulnerabilities in safety-aligned LLMs by eliciting harmful outputs through carefully crafted prompts. Existing methods rely on discrete optimization or trained adversarial generators, but are slow,…

Computation and Language · Computer Science 2025-07-08 James Beetham , Souradip Chakraborty , Mengdi Wang , Furong Huang , Amrit Singh Bedi , Mubarak Shah

Current safety alignment techniques for large language models (LLMs) face two key challenges: (1) under-generalization, which leaves models vulnerable to novel jailbreak attacks, and (2) over-alignment, which leads to the excessive refusal…

Computation and Language · Computer Science 2025-04-15 Yutao Mou , Yuxiao Luo , Shikun Zhang , Wei Ye

"Oh-Oh, yes, I'm the great pretender. Pretending that I'm doing well. My need is such, I pretend too much..." summarizes the state in the area of jailbreak creation and evaluation. You find this method to generate adversarial attacks…

Cryptography and Security · Computer Science 2026-05-15 Jean-Philippe Monteuuis , Cong Chen , Jonathan Petit

Large language models (LLMs) have made significant advancements across various tasks, but their safety alignment remain a major concern. Exploring jailbreak prompts can expose LLMs' vulnerabilities and guide efforts to secure them. Existing…

Cryptography and Security · Computer Science 2025-11-25 Xiaoning Dong , Wenbo Hu , Wei Xu , Tianxing He

As multimodal reasoning improves the overall capabilities of Large Vision Language Models (LVLMs), recent studies have begun to explore safety-oriented reasoning, aiming to enhance safety awareness by analyzing potential safety risks during…

Computer Vision and Pattern Recognition · Computer Science 2025-12-09 Fenghua Weng , Chaochao Lu , Xia Hu , Wenqi Shao , Wenjie Wang

Safety alignment in Large Language Models (LLMs) often creates a systematic discrepancy between a model's aligned output and the underlying pre-aligned data distribution. We propose a framework in which the effect of safety alignment on…

Computation and Language · Computer Science 2026-02-03 Yuxuan Lu , Yongkang Guo , Yuqing Kong

Alignment in large language models (LLMs) is used to enforce guidelines such as safety. Yet, alignment fails in the face of jailbreak attacks that modify inputs to induce unsafe outputs. In this paper, we introduce and evaluate a new…

Cryptography and Security · Computer Science 2026-02-19 Jean-Charles Noirot Ferrand , Yohan Beugin , Eric Pauley , Ryan Sheatsley , Patrick McDaniel

Large Language Models (LLMs) are powerful tools for answering user queries, yet they remain highly vulnerable to jailbreak attacks. Existing guardrail methods typically rely on internal features or textual responses to detect malicious…

Cryptography and Security · Computer Science 2026-05-29 Zikai Zhang , Rui Hu , Olivera Kotevska , Jiahao Xu

We use statistical tools to characterize the response of an excitable system to periodic perturbations. The system is an optically injected semiconductor laser under pulsed perturbations of the phase of the injected field. We characterize…

Adaptation and Self-Organizing Systems · Physics 2020-08-26 Jordi Tiana-Alsina , Bruno Garbin , Stephane Barland , Cristina Masoller

Automated LLM vulnerability scanners are increasingly used to assess security risks by measuring different attack type success rates (ASR). Yet the validity of these measurements hinges on an often-overlooked component: the evaluator who…

Cryptography and Security · Computer Science 2026-03-17 Lidor Erez , Omer Hofman , Tamir Nizri , Roman Vainshtein

Large Language Models (LLMs) are typically evaluated for safety under single-shot or low-budget adversarial prompting, which underestimates real-world risk. In practice, attackers can exploit large-scale parallel sampling to repeatedly…

Artificial Intelligence · Computer Science 2026-02-10 Mingqian Feng , Xiaodong Liu , Weiwei Yang , Chenliang Xu , Christopher White , Jianfeng Gao

Vision-language models (VLMs) have become central to tasks such as visual question answering, image captioning, and text-to-image generation. However, their outputs are highly sensitive to prompt variations, which can reveal vulnerabilities…

Computer Vision and Pattern Recognition · Computer Science 2026-02-02 Aarush Noheria , Yuguang Yao

Large language models remain vulnerable to jailbreak attacks, and single-layer defenses often trade security for usability. We present TRYLOCK, the first defense-in-depth architecture that combines four heterogeneous mechanisms across the…

Cryptography and Security · Computer Science 2026-01-08 Scott Thornton

It has been shown that natural language processing (NLP) models are vulnerable to a kind of security threat called the Backdoor Attack, which utilizes a `backdoor trigger' paradigm to mislead the models. The most threatening backdoor attack…

Computation and Language · Computer Science 2022-02-17 Lingfeng Shen , Haiyun Jiang , Lemao Liu , Shuming Shi

Safety-aligned LLMs respond to prompts with either compliance or refusal, each corresponding to distinct directions in the model's activation space. Recent works show that initializing attacks via self-transfer from other prompts…

Cryptography and Security · Computer Science 2025-10-09 Amit Levi , Rom Himelstein , Yaniv Nemcovsky , Avi Mendelson , Chaim Baskin

Despite rich safety alignment strategies, large language models (LLMs) remain highly susceptible to jailbreak attacks, which compromise safety guardrails and pose serious security risks. Existing detection methods mainly detect jailbreak…

Machine Learning · Computer Science 2026-01-08 Xiao Lin , Philip Li , Zhichen Zeng , Tingwei Li , Tianxin Wei , Xuying Ning , Gaotang Li , Yuzhong Chen , Hanghang Tong

Reasoning methods that adaptively allocate test-time compute have advanced LLM performance on easy to verify domains such as math and code. In this work, we study how to utilize this approach to train models that exhibit a degree of…

Machine Learning · Computer Science 2025-10-28 Taeyoun Kim , Fahim Tajwar , Aditi Raghunathan , Aviral Kumar

Reinforcement Learning (RL) has shown promise in various robotics applications, yet its deployment on real systems is still limited due to safety and operational constraints. The safe RL field has gained considerable attention in recent…

Robotics · Computer Science 2026-03-19 Sadık Bera Yüksel , Ali Tevfik Buyukkocak , Derya Aksaray

Jailbreak attacks against multimodal large language Models (MLLMs) are a significant research focus. Current research predominantly focuses on maximizing attack success rate (ASR), often overlooking whether the generated responses actually…