English
Related papers

Related papers: High-Entropy Tokens as Multimodal Failure Points i…

200 papers

Large Vision-Language Models (LVLMs) rely on attention-based retrieval of safety instructions to maintain alignment during generation. Existing attacks typically optimize image perturbations to maximize harmful output likelihood, but suffer…

Computer Vision and Pattern Recognition · Computer Science 2026-04-14 Jingru Li , Wei Ren , Tianqing Zhu

Balancing exploration and exploitation is a central goal in reinforcement learning (RL). Despite recent advances in enhancing large language model (LLM) reasoning, most methods lean toward exploitation, and increasingly encounter…

Computation and Language · Computer Science 2025-11-11 Daixuan Cheng , Shaohan Huang , Xuekai Zhu , Bo Dai , Wayne Xin Zhao , Zhenliang Zhang , Furu Wei

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a powerful approach to enhancing the reasoning capabilities of Large Language Models (LLMs), while its mechanisms are not yet well understood. In this work, we undertake a…

Despite substantial investment in safety alignment, the vulnerability of large language models to sophisticated multi-turn adversarial attacks remains poorly characterized, and whether model scale or inference mode affects robustness is…

Computation and Language · Computer Science 2025-12-09 Richard Young

Large Language Models (LLMs) are being increasingly deployed in real-world applications, but they remain susceptible to hallucinations, which produce fluent yet incorrect responses and lead to erroneous decision-making. Uncertainty…

Machine Learning · Computer Science 2025-12-02 Huan Ma , Jiadong Pan , Jing Liu , Yan Chen , Joey Tianyi Zhou , Guangyu Wang , Qinghua Hu , Hua Wu , Changqing Zhang , Haifeng Wang

Large Vision Language Models (LVLMs) excel in various vision-language tasks. Yet, their robustness to visual variations in position, scale, orientation, and context that objects in natural scenes inevitably exhibit due to changes in…

Computer Vision and Pattern Recognition · Computer Science 2025-06-03 Zhiyuan Fan , Yumeng Wang , Sandeep Polisetty , Yi R. Fung

Recent advances in Large Language Models (LLMs) have led to impressive alignment where models learn to distinguish harmful from harmless queries through supervised finetuning (SFT) and reinforcement learning from human feedback (RLHF). In…

Artificial Intelligence · Computer Science 2025-06-18 Jiahao Yu , Haozheng Luo , Jerry Yao-Chieh Hu , Wenbo Guo , Han Liu , Xinyu Xing

We introduce the Adversarial Confusion Attack, a new class of threats against multimodal large language models (MLLMs). Unlike jailbreaks or targeted misclassification, the goal is to induce systematic disruption that makes the model…

Computation and Language · Computer Science 2025-12-02 Jakub Hoscilowicz , Artur Janicki

Hallucinations remain a persistent challenge for vision-language models (VLMs), which often describe nonexistent objects or fabricate facts. Existing detection methods typically operate after text generation, making intervention both costly…

Computer Vision and Pattern Recognition · Computer Science 2026-03-06 Sai Akhil Kogilathota , Sripadha Vallabha E G , Luzhe Sun , Jiawei Zhou

Large vision-language models (LVLMs) often hallucinate content that is fluent yet unsupported by the image, limiting their reliability in real-world deployment. We show that a key failure mode arises from route competition: even when visual…

Computer Vision and Pattern Recognition · Computer Science 2026-05-26 Zhe Cheng , Wenyu Chen , Fode Zhang , Dehuan Shen

Vision language models (VLM) demonstrate sophisticated multimodal reasoning yet are prone to hallucination when confronted with knowledge conflicts, impeding their deployment in information-sensitive contexts. While existing research…

Computer Vision and Pattern Recognition · Computer Science 2025-07-18 Peter Carragher , Nikitha Rao , Abhinand Jha , R Raghav , Kathleen M. Carley

Large Language Models (LLMs) and Vision Language Models (VLMs) have demonstrated impressive capabilities but remain vulnerable to jailbreaking attacks, where adversaries exploit textual or visual triggers to bypass safety guardrails. Recent…

Cryptography and Security · Computer Science 2026-05-15 Yi Wang , Hongye Qiu , Yue Xu , Sibei Yang , Zhan Qin , Minlie Huang , Wenjie Wang

Vision-language pre-trained (VLP) models have been the foundation of numerous vision-language tasks. Given their prevalence, it becomes imperative to assess their adversarial robustness, especially when deploying them in security-crucial…

Computer Vision and Pattern Recognition · Computer Science 2024-05-12 Peng-Fei Zhang , Zi Huang , Guangdong Bai

Adversarial images pose a severe security threat to multimodal large language models through prompt injection. Existing defenses largely lack a principled understanding of the underlying mechanisms and struggle to balance efficiency and…

Machine Learning · Computer Science 2026-05-26 Dongpeng Zhang , Ke Ma , Yangbangyan Jiang , Gaozheng Pei , Longtao Huang , Qianqian Xu , Qingming Huang

Pretrained vision-language models (VLMs) like CLIP exhibit exceptional generalization across diverse downstream tasks. While recent studies reveal their vulnerability to adversarial attacks, research to date has primarily focused on…

Computer Vision and Pattern Recognition · Computer Science 2024-11-13 Wanqi Zhou , Shuanghao Bai , Danilo P. Mandic , Qibin Zhao , Badong Chen

Despite the rapid progress in multimodal models and Large Visual-Language Models (LVLM), they remain highly susceptible to adversarial perturbations, raising serious concerns about their reliability in real-world use. While adversarial…

Computer Vision and Pattern Recognition · Computer Science 2026-04-01 Mujtaba Hussain Mirza , Antonio D'Orazio , Odelia Melamed , Iacopo Masi

Large language models (LLMs) have significantly influenced various industries but suffer from a critical flaw, the potential sensitivity of generating harmful content, which poses severe societal risks. We developed and tested novel attack…

Computation and Language · Computer Science 2025-02-25 Yuyi Huang , Runzhe Zhan , Derek F. Wong , Lidia S. Chao , Ailin Tao

The objective of drug discovery is to identify chemical compounds that possess specific pharmaceutical properties toward a binding target. Existing large language models (LLMS) can achieve high token matching scores in terms of likelihood…

Machine Learning · Computer Science 2025-04-01 Xuefeng Liu , Chih-chan Tien , Peng Ding , Songhao Jiang , Rick L. Stevens

Vision-Language Models (VLMs) are increasingly deployed in real-world applications, but their high inference cost makes them vulnerable to resource consumption attacks. Prior attacks attempt to extend VLM output sequences by optimizing…

Cryptography and Security · Computer Science 2025-08-27 Rui Zhang , Zihan Wang , Tianli Yang , Hongwei Li , Wenbo Jiang , Qingchuan Zhao , Yang Liu , Guowen Xu

Vision-Language Models (VLMs) have demonstrated strong capability in a wide range of tasks such as visual recognition, document parsing, and visual grounding. Nevertheless, recent work shows that while VLMs often manage to capture the…

Computer Vision and Pattern Recognition · Computer Science 2026-04-20 Chengxin Liu , Wonseok Choi , Chenshuang Zhang , Tae-Hyun Oh