English
Related papers

Related papers: Consistent but Dangerous: Per-Sample Safety Classi…

200 papers

Although language model scores are often treated as probabilities, their reliability as probability estimators has mainly been studied through calibration, overlooking other aspects. In particular, it is unclear whether language models…

Computation and Language · Computer Science 2024-10-01 Eitan Wagner , Yuli Slavutsky , Omri Abend

Typographic prompt injection exploits vision language models' (VLMs) ability to read text rendered in images, posing a growing threat as VLMs power autonomous agents. Prior work typically focus on maximizing attack success rate (ASR) but…

Computer Vision and Pattern Recognition · Computer Science 2026-04-29 Ravikumar Balakrishnan , Sanket Mendapara

Training large language models to follow instructions makes them perform better on a wide range of tasks and generally become more helpful. However, a perfectly helpful model will follow even the most malicious instructions and readily…

Computation and Language · Computer Science 2024-03-20 Federico Bianchi , Mirac Suzgun , Giuseppe Attanasio , Paul Röttger , Dan Jurafsky , Tatsunori Hashimoto , James Zou

The emergence of Vision Language Models (VLMs) has brought unprecedented advances in understanding multimodal information. The combination of textual and visual semantics in VLMs is highly complex and diverse, making the safety alignment of…

Computer Vision and Pattern Recognition · Computer Science 2025-05-22 Yongting Zhang , Lu Chen , Guodong Zheng , Yifeng Gao , Rui Zheng , Jinlan Fu , Zhenfei Yin , Senjie Jin , Yu Qiao , Xuanjing Huang , Feng Zhao , Tao Gui , Jing Shao

Vision-Language Models (VLMs) have great potential in medical tasks, like Visual Question Answering (VQA), where they could act as interactive assistants for both patients and clinicians. Yet their robustness to distribution shifts on…

Computer Vision and Pattern Recognition · Computer Science 2025-07-04 Kim-Celine Kahl , Selen Erkan , Jeremias Traub , Carsten T. Lüth , Klaus Maier-Hein , Lena Maier-Hein , Paul F. Jaeger

The robust safety of Vision-Language Large Models (VLLMs) against joint multilingual and multimodal threats remains severely underexplored. Current benchmarks typically isolate these dimensions, being either multilingual but text-only, or…

Computer Vision and Pattern Recognition · Computer Science 2026-04-22 Enyi Shi , Pengyang Shao , Yanxin Zhang , Chenhang Cui , Jiayi Lyu , Xiaobo Xia , Fei Shen , Tat-Seng Chua

Medical vision--language models (VLMs) are usually evaluated on intact image--question pairs, but trustworthy clinical use requires a stronger property: a model must recognise when the evidential basis for an answer has failed. We study…

Computer Vision and Pattern Recognition · Computer Science 2026-05-25 Hanqi Jiang , Junhao Chen , Mingyu Kang , Hyeokjae Kwon , Yi Pan , Lifeng Chen , Weihang You , Haozhen Gong , Ruiyu Yan , Jinglei Lv , Lin Zhao , Hui Ren , Quanzheng Li , Tianming Liu , Xiang Li

Vision-Language Models (VLMs) have achieved impressive performance across a wide range of multimodal tasks, yet they often exhibit inconsistent behavior when faced with semantically equivalent inputs, undermining their reliability and…

Computer Vision and Pattern Recognition · Computer Science 2025-06-30 Shih-Han Chou , Shivam Chandhok , James J. Little , Leonid Sigal

Medical vision-language models (VLMs) offer promise for clinical decision support, yet their reliability under distribution shifts remains a major concern for safe deployment. These models often learn task-agnostic correlations due to…

Computer Vision and Pattern Recognition · Computer Science 2025-09-12 Umaima Rahman , Raza Imam , Mohammad Yaqub , Dwarikanath Mahapatra

While large language models (LLMs) such as Llama-2 or GPT-4 have shown impressive zero-shot performance, fine-tuning is still necessary to enhance their performance for customized datasets, domain-specific tasks, or other private needs.…

Machine Learning · Computer Science 2025-01-07 Chia-Yi Hsu , Yu-Lin Tsai , Chih-Hsun Lin , Pin-Yu Chen , Chia-Mu Yu , Chun-Ying Huang

Clinical LLMs are often scaled by increasing model size, context length, retrieval complexity, or inference-time compute, with the implicit expectation that higher accuracy implies safer behavior. This assumption is incomplete in medicine,…

Vision-Language Models (VLMs) have demonstrated strong capabilities in aligning visual and textual modalities, enabling a wide range of applications in multimodal understanding and generation. While they excel in zero-shot and transfer…

Computer Vision and Pattern Recognition · Computer Science 2025-09-25 Hao Dong , Moru Liu , Jian Liang , Eleni Chatzi , Olga Fink

Large language models require continuous adaptation to new tasks while preserving safety alignment. However, fine-tuning on even benign data often compromises safety behaviors, including refusal of harmful requests, truthfulness, and…

Machine Learning · Computer Science 2026-04-21 Thong Bach , Dung Nguyen , Thao Minh Le , Truyen Tran

Vision-language models (VLMs) are increasingly applied to identify unsafe or inappropriate images due to their internal ethical standards and powerful reasoning abilities. However, it is still unclear whether they can recognize various…

Cryptography and Security · Computer Science 2025-07-16 Yiting Qu , Michael Backes , Yang Zhang

As vision-language models (VLMs) become increasingly capable, maintaining a balance between safety and usefulness remains a central challenge. Safety mechanisms, while essential, can backfire, causing over-refusal, where models decline…

Computation and Language · Computer Science 2026-03-20 Kaixuan Ren , Preslav Nakov , Usman Naseem

Consistency, which refers to the capability of generating the same predictions for semantically similar contexts, is a highly desirable property for a sound language understanding model. Although recent pretrained language models (PLMs)…

Computation and Language · Computer Science 2021-08-17 Myeongjun Jang , Deuk Sin Kwon , Thomas Lukasiewicz

Safety evaluations of language models often treat serving configuration as fixed background infrastructure, but batch condition is an untested treatment variable whenever the same prompt may be evaluated alone, in a synchronized batch, or…

Machine Learning · Computer Science 2026-05-28 Sahil Kadadekar

Fine-tuning language models is commonly believed to inevitably harm their safety, i.e., refusing to respond to harmful user requests, even when using harmless datasets, thus requiring additional safety measures. We challenge this belief…

Machine Learning · Computer Science 2025-08-19 Minseon Kim , Jin Myung Kwak , Lama Alssum , Bernard Ghanem , Philip Torr , David Krueger , Fazl Barez , Adel Bibi

Fine-tuning safety-aligned large language models (LLMs) can substantially compromise their safety. Previous approaches require many safety samples or calibration sets, which not only incur significant computational overhead during…

Machine Learning · Computer Science 2026-01-07 Jiawen Zhang , Lipeng He , Kejia Chen , Jian Lou , Jian Liu , Xiaohu Yang , Ruoxi Jia

A pervasive intuition holds that vision-language models (VLMs) are most trustworthy when their attention maps look sharp: concentrated attention on the queried region should imply a confident, calibrated answer. We test this…

Artificial Intelligence · Computer Science 2026-05-12 Logan Mann , Ajit Saravanan , Ishan Dave , Shikhar Shiromani , Saadullah Ismail , Yi Xia , Emily Huang