English
Related papers

Related papers: Robust LLM safeguarding via refusal feature advers…

200 papers

The growing misuse of Vision-Language Models (VLMs) has led providers to deploy multiple safeguards, including alignment tuning, system prompts, and content moderation. However, the real-world robustness of these defenses against…

Cryptography and Security · Computer Science 2025-11-21 Yijun Yang , Lichao Wang , Jianping Zhang , Chi Harold Liu , Lanqing Hong , Qiang Xu

Large Language Models (LLMs) represent a transformative leap in artificial intelligence, enabling the comprehension, generation, and nuanced interaction with human language on an unparalleled scale. However, LLMs are increasingly vulnerable…

Cryptography and Security · Computer Science 2025-02-06 Nan Wang , Kane Walter , Yansong Gao , Alsharif Abuadbba

Despite extensive diagnostics and debugging by developers, AI systems sometimes exhibit harmful unintended behaviors. Finding and fixing these is challenging because the attack surface is so large -- it is not tractable to exhaustively…

Cryptography and Security · Computer Science 2025-07-30 Stephen Casper , Lennart Schulze , Oam Patel , Dylan Hadfield-Menell

Adversarial training in reinforcement learning (RL) is challenging because perturbations cascade through trajectories and compound over time, making fixed-strength attacks either overly destructive or too conservative. We propose…

Machine Learning · Computer Science 2026-01-30 Lucas Schott , Elies Gherbi , Hatem Hajri , Sylvain Lamprier

Reinforcement Learning with Human Feedback (RLHF) has been demonstrated to significantly enhance the performance of large language models (LLMs) by aligning their outputs with desired human values through instruction tuning. However, RLHF…

Computation and Language · Computer Science 2024-03-06 Zhang Ze Yu , Lau Jia Jaw , Zhang Hui , Bryan Kian Hsiang Low

Function call capabilities have become crucial for Large Language Models (LLMs), enabling them to interact more effectively with external tools and APIs. Existing methods for improving the function call capabilities of LLMs rely on data…

Artificial Intelligence · Computer Science 2026-01-28 Weiran Guo , Bing Bo , Shaoxiang Wu , Jingsheng Yang

Large Audio Language Models (LALMs) have extended the capabilities of Large Language Models (LLMs) by enabling audio-based human interactions. However, recent research has revealed that LALMs remain vulnerable to harmful queries due to…

Computation and Language · Computer Science 2025-05-27 Hao Yang , Lizhen Qu , Ehsan Shareghi , Gholamreza Haffari

Large language models (LLMs) remain vulnerable to adversarial prompting despite advances in alignment and safety, often exhibiting harmful behaviors under novel attack strategies. While adversarial training can improve robustness, existing…

Machine Learning · Computer Science 2026-05-08 Yiwei Zhang , Jeremiah Birrell , Reza Ebrahimi , Rouzbeh Behnia , Jason Pacheco , Elisa Bertino

The release of open-weight large language models (LLMs) creates a tension between advancing accessible research and preventing misuse, such as malicious fine-tuning to elicit harmful content. Current safety measures struggle to preserve the…

Computation and Language · Computer Science 2025-09-11 Debdeep Sanyal , Manodeep Ray , Murari Mandal

We identify a structural weakness in current large language model (LLM) alignment: modern refusal mechanisms are fail-open. While existing approaches encode refusal behaviors across multiple latent features, suppressing a single dominant…

Machine Learning · Computer Science 2026-02-20 Zachary Coalson , Beth Sohler , Aiden Gabriel , Sanghyun Hong

Large Language Models (LLMs) exhibit substantial capabilities yet encounter challenges, including hallucination, outdated knowledge, and untraceable reasoning processes. Retrieval-augmented generation (RAG) has emerged as a promising…

Artificial Intelligence · Computer Science 2024-06-03 Feiteng Fang , Yuelin Bai , Shiwen Ni , Min Yang , Xiaojun Chen , Ruifeng Xu

As powerful Large Language Models (LLMs) are now widely used for numerous practical applications, their safety is of critical importance. While alignment techniques have significantly improved overall safety, LLMs remain vulnerable to…

Machine Learning · Computer Science 2024-10-28 Samuel Jacob Chacko , Sajib Biswas , Chashi Mahiul Islam , Fatema Tabassum Liza , Xiuwen Liu

Recently, Large Language Models (LLMs) have made significant advancements and are now widely used across various domains. Unfortunately, there has been a rising concern that LLMs can be misused to generate harmful or malicious content.…

Computation and Language · Computer Science 2024-06-13 Bochuan Cao , Yuanpu Cao , Lu Lin , Jinghui Chen

Large language model (LLM) safety classifiers such as Llama Guard are effective at detecting overtly harmful prompts but remain vulnerable to adversarial jailbreak attacks that disguise malicious intent through role-play scenarios,…

Cryptography and Security · Computer Science 2026-05-26 Lixing Lin , Juli You , Yue Li , Luyun Lin , Yiqing Wang , Zhen Zhang , Moxuan Zheng

Reinforcement learning algorithms, just like any other Machine learning algorithm pose a serious threat from adversaries. The adversaries can manipulate the learning algorithm resulting in non-optimal policies. In this paper, we analyze the…

Machine Learning · Computer Science 2021-03-12 Aqeel Anwar , Arijit Raychowdhury

Deep learning models continue to advance in accuracy, yet they remain vulnerable to adversarial attacks, which often lead to the misclassification of adversarial examples. Adversarial training is used to mitigate this problem by increasing…

Computer Vision and Pattern Recognition · Computer Science 2024-02-20 Leo Hyun Park , Jaeuk Kim , Myung Gyo Oh , Jaewoo Park , Taekyoung Kwon

Large Language Models (LLMs) are valuable for text classification, but their vulnerabilities must not be disregarded. They lack robustness against adversarial examples, so it is pertinent to understand the impacts of different types of…

Computation and Language · Computer Science 2024-06-13 João Vitorino , Eva Maia , Isabel Praça

To circumvent the alignment of large language models (LLMs), current optimization-based adversarial attacks usually craft adversarial prompts by maximizing the likelihood of a so-called affirmative response. An affirmative response is a…

Adversarial training (AT) is an effective defense for large language models (LLMs) against jailbreak attacks, but performing AT on LLMs is costly. To improve the efficiency of AT for LLMs, recent studies propose continuous AT (CAT) that…

Machine Learning · Computer Science 2026-04-15 Shaopeng Fu , Di Wang

Robust verbal confidence generated by large language models (LLMs) is crucial for the deployment of LLMs to help ensure transparency, trust, and safety in many applications, including those involving human-AI interactions. In this paper, we…

Computation and Language · Computer Science 2025-12-19 Stephen Obadinma , Xiaodan Zhu