English
Related papers

Related papers: Refusal in Language Models Is Mediated by a Single…

200 papers

We introduce Directional Stimulus Prompting, a novel framework for guiding black-box large language models (LLMs) toward specific desired outputs. Instead of directly adjusting LLMs, our method employs a small tunable policy model (e.g.,…

Computation and Language · Computer Science 2023-10-11 Zekun Li , Baolin Peng , Pengcheng He , Michel Galley , Jianfeng Gao , Xifeng Yan

Large Language Models (LLMs) have demonstrated exceptional capabilities across various natural language processing tasks. Due to their training on internet-sourced datasets, LLMs can sometimes generate objectionable content, necessitating…

Computation and Language · Computer Science 2024-11-15 Leyang Hu , Boran Wang

The rapid progress of Large Language Models (LLMs) has opened up new opportunities across various domains and applications; yet it also presents challenges related to potential misuse. To mitigate such risks, red teaming has been employed…

Cryptography and Security · Computer Science 2025-06-10 Yifan Jiang , Kriti Aggarwal , Tanmay Laud , Kashif Munir , Jay Pujara , Subhabrata Mukherjee

Extensive work has been devoted to improving the safety mechanism of Large Language Models (LLMs). However, LLMs still tend to generate harmful responses when faced with malicious instructions, a phenomenon referred to as "Jailbreak…

Computation and Language · Computer Science 2024-02-26 Yanrui Du , Sendong Zhao , Ming Ma , Yuhan Chen , Bing Qin

Current safety alignment of foundation models largely follows a \emph{one-size-fits-all} paradigm, applying the same refusal policy across users and contexts. As a result, models may refuse requests that are unsafe for general users but…

Artificial Intelligence · Computer Science 2026-05-26 Qitao Tan , Xiaoying Song , Arman Akbari , Arash Akbari , Yanzhi Wang , Xiaoming Zhai , Lingzi Hong , Zhen Xiang , Jin Lu , Geng Yuan

Large (vision-)language models exhibit remarkable capability but remain highly susceptible to jailbreaking. Existing safety training approaches aim to have the model learn a refusal boundary between safe and unsafe, based on the user's…

Cryptography and Security · Computer Science 2026-04-28 Xinhe Wang , Katia Sycara , Yaqi Xie

Large language models are finetuned to refuse questions about hazardous knowledge, but these protections can often be bypassed. Unlearning methods aim at completely removing hazardous capabilities from models and make them inaccessible to…

Machine Learning · Computer Science 2025-06-03 Jakub Łucki , Boyi Wei , Yangsibo Huang , Peter Henderson , Florian Tramèr , Javier Rando

Large language models (LLMs) are increasingly deployed in medical contexts, raising critical concerns about safety, alignment, and susceptibility to adversarial manipulation. While prior benchmarks assess model refusal capabilities for…

Computation and Language · Computer Science 2025-05-19 Sijia Chen , Xiaomin Li , Mengxue Zhang , Eric Hanchen Jiang , Qingcheng Zeng , Chen-Hsiang Yu

This work introduces a novel framework for evaluating LLMs' capacity to balance instruction-following with critical reasoning when presented with multiple-choice questions containing no valid answers. Through systematic evaluation across…

Computation and Language · Computer Science 2025-06-03 Gracjan Góral , Emilia Wiśnios , Piotr Sankowski , Paweł Budzianowski

Large Language Models (LLMs) like OpenAI's GPT series, Anthropic's Claude, and Meta's LLaMa have shown remarkable capabilities in text generation. However, their susceptibility to toxic prompts presents significant security challenges. This…

Cryptography and Security · Computer Science 2024-12-03 Jie Li , Yi Liu , Chongyang Liu , Xiaoning Ren , Ling Shi , Weisong Sun , Yinxing Xue

Protecting the intellectual property of large language models (LLMs) is a critical challenge due to the proliferation of unauthorized derivative models. We introduce a novel fingerprinting framework that leverages the behavioral patterns…

Cryptography and Security · Computer Science 2026-02-11 Zhenyu Xu , Victor S. Sheng

Large language models (LLMs) excel in various tasks but remain vulnerable to jailbreak attacks, where adversaries manipulate prompts to generate harmful outputs. Examining jailbreak prompts helps uncover the shortcomings of LLMs. However,…

Computation and Language · Computer Science 2024-12-18 Weixiong Zheng , Peijian Zeng , Yiwei Li , Hongyan Wu , Nankai Lin , Junhao Chen , Aimin Yang , Yongmei Zhou

Safety-aligned language models often refuse prompts that are actually harmless. Current evaluations mostly report global rates such as false rejection or compliance. These scores treat each prompt alone and miss local inconsistency, where a…

Computation and Language · Computer Science 2025-12-22 Riad Ahmed Anonto , Md Labid Al Nahiyan , Md Tanvir Hassan

In this study, we disclose a worrying new vulnerability in Large Language Models (LLMs), which we term \textbf{involuntary jailbreak}. Unlike existing jailbreak attacks, this weakness is distinct in that it does not involve a specific…

Cryptography and Security · Computer Science 2025-12-30 Yangyang Guo , Yangyan Li , Mohan Kankanhalli

The increasing integration of large language models (LLMs) into mental health applications necessitates robust frameworks for evaluating professional safety alignment. Current evaluative approaches primarily rely on refusal-based safety…

Computation and Language · Computer Science 2026-01-08 Yaling Shen , Stephanie Fong , Yiwen Jiang , Zimu Wang , Feilong Tang , Qingyang Xu , Xiangyu Zhao , Zhongxing Xu , Jiahe Liu , Jinpeng Hu , Dominic Dwyer , Zongyuan Ge

Neural dialogue models have been widely adopted in various chatbot applications because of their good performance in simulating and generalizing human conversations. However, there exists a dark side of these models -- due to the…

Computation and Language · Computer Science 2019-09-30 Haochen Liu , Tyler Derr , Zitao Liu , Jiliang Tang

Safety-aligned language models often refuse cybersecurity requests whose wording resembles misuse, even when the task is authorized and defensive. This makes security evaluation ambiguous: a failed answer may reflect missing capability or…

Cryptography and Security · Computer Science 2026-05-19 Isaac David , Arthur Gervais

Refusal behavior in aligned LLMs is often viewed as model-specific, yet we hypothesize it stems from a universal, low-dimensional semantic circuit shared across models. To test this, we introduce Trajectory Replay via Concept-Basis…

Computation and Language · Computer Science 2026-01-27 Tony Cristofano

Instruction Fine-Tuning (IFT) has been widely adopted as an effective post-training strategy to enhance various abilities of Large Language Models (LLMs). However, prior studies have shown that IFT can significantly compromise LLMs' safety,…

Computation and Language · Computer Science 2025-09-09 Yanrui Du , Fenglei Fan , Sendong Zhao , Jiawei Cao , Qika Lin , Kai He , Ting Liu , Bing Qin , Mengling Feng

We identify a novel phenomenon in language models: benign fine-tuning of frontier models can lead to privacy collapse. We find that diverse, subtle patterns in training data can degrade contextual privacy, including optimisation for…

Computation and Language · Computer Science 2026-04-21 Anmol Goel , Cornelius Emde , Sangdoo Yun , Seong Joon Oh , Martin Gubri