English
Related papers

Related papers: A Single Neuron Is Sufficient to Bypass Safety Ali…

200 papers

AI systems can take harmful actions and are highly vulnerable to adversarial attacks. We present an approach, inspired by recent advances in representation engineering, that interrupts the models as they respond with harmful outputs with…

Safety alignment in Large Language Models (LLMs) remains highly fragile during fine-tuning, where even benign adaptation can degrade pre-trained refusal behaviors and enable harmful responses. Existing defenses typically constrain either…

Artificial Intelligence · Computer Science 2026-04-15 Songping Peng , Zhiheng Zhang , Daojian Zeng , Lincheng Jiang , Xieping Gao

Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of tasks, yet they often refuse to answer legitimate queries--a phenomenon known as overrefusal. Overrefusal typically stems from over-conservative…

Artificial Intelligence · Computer Science 2025-09-18 Licheng Pan , Yongqi Tong , Xin Zhang , Xiaolu Zhang , Jun Zhou , Zhixuan Chu

A core challenge in the development of increasingly capable AI systems is to make them safe and reliable by ensuring their behaviour is consistent with human values. This challenge, known as the alignment problem, does not merely apply to…

Machine Learning · Computer Science 2023-11-07 Raphaël Millière

Large reasoning models with reasoning capabilities achieve state-of-the-art performance on complex tasks, but their robustness under multi-turn adversarial pressure remains underexplored. We evaluate nine frontier reasoning models under…

Artificial Intelligence · Computer Science 2026-03-13 Yubo Li , Ramayya Krishnan , Rema Padman

Recent advancements in Large Language Models (LLMs) have sparked widespread concerns about their safety. Recent work demonstrates that safety alignment of LLMs can be easily removed by fine-tuning with a few adversarially chosen…

Computation and Language · Computer Science 2025-03-03 Samuele Poppi , Zheng-Xin Yong , Yifei He , Bobbie Chern , Han Zhao , Aobo Yang , Jianfeng Chi

The safety alignment of current Large Language Models (LLMs) is vulnerable. Relatively simple attacks, or even benign fine-tuning, can jailbreak aligned models. We argue that many of these vulnerabilities are related to a shared underlying…

Cryptography and Security · Computer Science 2024-06-11 Xiangyu Qi , Ashwinee Panda , Kaifeng Lyu , Xiao Ma , Subhrajit Roy , Ahmad Beirami , Prateek Mittal , Peter Henderson

Large Language Models (LLMs) have been extensively used across diverse domains, including virtual assistants, automated code generation, and scientific research. However, they remain vulnerable to jailbreak attacks, which manipulate the…

Cryptography and Security · Computer Science 2026-01-05 Haoran Gu , Handing Wang , Yi Mei , Mengjie Zhang , Yaochu Jin

Large language models (LLMs) undergo safety alignment to ensure safe conversations with humans. However, this paper introduces a training-free attack method capable of reversing safety alignment, converting the outcomes of stronger…

Computation and Language · Computer Science 2024-06-07 Zhanhui Zhou , Jie Liu , Zhichen Dong , Jiaheng Liu , Chao Yang , Wanli Ouyang , Yu Qiao

Generalizable alignment is a core challenge for deploying Large Language Models (LLMs) safely in real-world NLP applications. Current alignment methods, including Reinforcement Learning from Human Feedback (RLHF), often fail to guarantee…

Computation and Language · Computer Science 2025-04-07 Jaymari Chua , Chen Wang , Lina Yao

Large language models (LLMs) have emerged as powerful tools for addressing a wide range of general inquiries and tasks. Despite this, fine-tuning aligned LLMs on smaller, domain-specific datasets, critical to adapting them to specialized…

Artificial Intelligence · Computer Science 2025-02-04 Guanlin Li , Kangjie Chen , Shangwei Guo , Jie Zhang , Han Qiu , Chao Zhang , Guoyin Wang , Tianwei Zhang , Jiwei Li

Targeted syntactic evaluations have demonstrated the ability of language models to perform subject-verb agreement given difficult contexts. To elucidate the mechanisms by which the models accomplish this behavior, this study applies causal…

Computation and Language · Computer Science 2021-06-23 Matthew Finlayson , Aaron Mueller , Sebastian Gehrmann , Stuart Shieber , Tal Linzen , Yonatan Belinkov

Large language models (LLMs) undergo alignment training to avoid harmful behaviors, yet the resulting safeguards remain brittle: jailbreaks routinely bypass them, and fine-tuning on narrow domains can induce ``emergent misalignment'' that…

Computation and Language · Computer Science 2026-04-13 Hadas Orgad , Boyi Wei , Kaden Zheng , Martin Wattenberg , Peter Henderson , Seraphina Goldfarb-Tarrant , Yonatan Belinkov

Warning: This paper contains examples of harmful language, and reader discretion is recommended. The increasing open release of powerful large language models (LLMs) has facilitated the development of downstream applications by reducing the…

Computation and Language · Computer Science 2023-10-05 Xianjun Yang , Xiao Wang , Qi Zhang , Linda Petzold , William Yang Wang , Xun Zhao , Dahua Lin

As large language models (LLMs) become easily accessible nowadays, the trade-off between safety and helpfulness can significantly impact user experience. A model that prioritizes safety will cause users to feel less engaged and assisted…

Computation and Language · Computer Science 2024-04-02 Yi-Lin Tuan , Xilun Chen , Eric Michael Smith , Louis Martin , Soumya Batra , Asli Celikyilmaz , William Yang Wang , Daniel M. Bikel

Post-training alignment optimizes language models to match human preference signals, but this objective is not equivalent to modeling observed human behavior. We compare 120 base-aligned model pairs on more than 10,000 real human decisions…

Computation and Language · Computer Science 2026-05-27 Eilam Shapira , Moshe Tennenholtz , Roi Reichart

Large language models are meticulously aligned to be both helpful and harmless. However, recent research points to a potential overkill which means models may refuse to answer benign queries. In this paper, we investigate the factors for…

Computation and Language · Computer Science 2024-02-01 Chenyu Shi , Xiao Wang , Qiming Ge , Songyang Gao , Xianjun Yang , Tao Gui , Qi Zhang , Xuanjing Huang , Xun Zhao , Dahua Lin

This paper presents a systematic evaluation of Large Language Models' (LLMs) behavior on long-tail distributed (encrypted) texts and their safety implications. We introduce a two-dimensional framework for assessing LLM safety: (1)…

Computation and Language · Computer Science 2025-06-05 Utsav Maskey , Mark Dras , Usman Naseem

Large language models increasingly rely on explicit chain-of-thought reasoning to solve complex tasks, yet the safety of the reasoning process itself remains largely unaddressed. Existing work focuses predominantly on content safety (i.e.,…

Artificial Intelligence · Computer Science 2026-05-07 Xunguang Wang , Yuguang Zhou , Qingyue Wang , Zongjie Li , Ruixuan Huang , Zhenlan Ji , Pingchuan Ma , Shuai Wang

As large reasoning models (LRMs) grow more capable, chain-of-thought (CoT) reasoning introduces new safety challenges. Existing SFT-based safety alignment studies dominantly focused on filtering prompts with safe, high-quality responses,…

Computation and Language · Computer Science 2026-03-31 Raj Vardhan Tomar , Preslav Nakov , Yuxia Wang
‹ Prev 1 4 5 6 7 8 10 Next ›