English
Related papers

Related papers: Attention Is Where You Attack

200 papers

Despite the high quality performance of the deep neural network in real-world applications, they are susceptible to minor perturbations of adversarial attacks. This is mostly undetectable to human vision. The impact of such attacks has…

Computer Vision and Pattern Recognition · Computer Science 2021-01-18 K Naveen Kumar , C Vishnu , Reshmi Mitra , C Krishna Mohan

While modern text-to-image models excel at prompt-based generation, they often lack the fine-grained control necessary for specific user requirements like spatial layouts or subject appearances. Multi-condition control addresses this, yet…

Computer Vision and Pattern Recognition · Computer Science 2026-02-09 Chao Zhou , Tianyi Wei , Yiling Chen , Wenbo Zhou , Nenghai Yu

This study provides a new understanding of the adversarial attack problem by examining the correlation between adversarial attack and visual attention change. In particular, we observed that: (1) images with incomplete attention regions are…

Computer Vision and Pattern Recognition · Computer Science 2023-01-02 Shangxi Wu , Jitao Sang , Kaiyuan Xu , Jiaming Zhang , Jian Yu

Existing data poisoning attacks on retrieval-augmented generation (RAG) systems scale poorly because they require costly optimization of poisoned documents for each target phrase. We introduce Eyes-on-Me, a modular attack that decomposes an…

Machine Learning · Computer Science 2025-12-09 Yen-Shan Chen , Sian-Yao Huang , Cheng-Lin Yang , Yun-Nung Chen

Large Reasoning Models (LRMs) have achieved tremendous success with their chain-of-thought (CoT) reasoning, yet also face safety issues similar to those of basic language models. In particular, while algorithms are designed to guide them to…

Machine Learning · Computer Science 2026-02-05 Zeming Wei , Qiaosheng Zhang , Xia Hu , Xingcheng Xu

When building Large Language Models (LLMs), it is paramount to bear safety in mind and protect them with guardrails. Indeed, LLMs should never generate content promoting or normalizing harmful, illegal, or unethical behavior that may…

Computation and Language · Computer Science 2024-06-25 Simone Tedeschi , Felix Friedrich , Patrick Schramowski , Kristian Kersting , Roberto Navigli , Huu Nguyen , Bo Li

We present a multi-agent learning algorithm, ALMA-Learning, for efficient and fair allocations in large-scale systems. We circumvent the traditional pitfalls of multi-agent learning (e.g., the moving target problem, the curse of…

Multiagent Systems · Computer Science 2021-06-22 Panayiotis Danassis , Florian Wiedemair , Boi Faltings

The vulnerability of the high-performance machine learning models implies a security risk in applications with real-world consequences. Research on adversarial attacks is beneficial in guiding the development of machine learning models on…

Machine Learning · Computer Science 2022-11-16 Yiran Huang , Yexu Zhou , Michael Hefenbrock , Till Riedel , Likun Fang , Michael Beigl

In the past few years, Language Models (LMs) have shown par-human capabilities in several domains. Despite their practical applications and exceeding user consumption, they are susceptible to jailbreaks when malicious input exploits the…

Computation and Language · Computer Science 2025-04-18 Charlotte Siska , Anush Sankaran

Most researchers have tried to enhance the robustness of DNNs by revealing and repairing the vulnerability of DNNs with specialized adversarial examples. Parts of the attack examples have imperceptible perturbations restricted by Lp norm.…

Computer Vision and Pattern Recognition · Computer Science 2024-05-29 Yihao Huang , Liangru Sun , Qing Guo , Felix Juefei-Xu , Jiayi Zhu , Jincao Feng , Yang Liu , Geguang Pu

Effective attribution of Advanced Persistent Threats (APTs) increasingly hinges on the ability to correlate behavioral patterns and reason over complex, varied threat intelligence artifacts. We present AURA (Attribution Using…

Cryptography and Security · Computer Science 2025-06-13 Nanda Rani , Sandeep Kumar Shukla

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

Reinforcement learning (RL) fine-tuning transforms large language models while creating a vulnerability we experimentally verify: Our experiment shows that malicious RL fine-tuning dismantles safety guardrails with remarkable efficiency,…

Machine Learning · Computer Science 2025-05-08 Wenjun Cao

Attention regulates information transfer between tokens. For this, query and key vectors are compared, typically in terms of a scalar product, $\mathbf{Q}^T\mathbf{K}$, together with a subsequent softmax normalization. In geometric terms,…

Machine Learning · Computer Science 2025-08-11 Claudius Gros

Reasoning methods that adaptively allocate test-time compute have advanced LLM performance on easy to verify domains such as math and code. In this work, we study how to utilize this approach to train models that exhibit a degree of…

Machine Learning · Computer Science 2025-10-28 Taeyoun Kim , Fahim Tajwar , Aditi Raghunathan , Aviral Kumar

Safety alignment in large language models (LLMs) is increasingly compromised by jailbreak attacks, which can manipulate these models to generate harmful or unintended content. Investigating these attacks is crucial for uncovering model…

Cryptography and Security · Computer Science 2025-05-26 Linbao Li , Yannan Liu , Daojing He , Yu Li

Even modern AI models often remain vulnerable to multimodal queries in which harmful intent is embedded in images. A widely used approach for safety alignment is training with extensive multimodal safety datasets, but the costs of data…

Computer Vision and Pattern Recognition · Computer Science 2026-03-30 Jonghyun Park , Minhyuk Seo , Chaewon Yeo , Jonghyun Choi

To address the trade-off between robustness and performance for robust VLM, we observe that function words could incur vulnerability of VLMs against cross-modal adversarial attacks, and propose Function-word De-Attention (FDA) accordingly…

Machine Learning · Computer Science 2026-04-17 Qiwei Tian , Chenhao Lin , Zhengyu Zhao , Chao Shen

A popular class of defenses against prompt injection attacks on large language models (LLMs) relies on fine-tuning to separate instructions and data, so that the LLM does not follow instructions that might be present with data. We evaluate…

Cryptography and Security · Computer Science 2025-12-18 Nishit V. Pandya , Andrey Labunets , Sicun Gao , Earlence Fernandes

Diffusion Language Models (DLMs) enable globally coherent, bidirectional, and controllable text generation, offering advantages over traditional autoregressive LLMs, while scaling to ultra-long sequences remains costly. Many existing…

Computer Vision and Pattern Recognition · Computer Science 2026-05-20 Wenhu Zhang , Yiming Wu , Huanyu Wang , Yaoyang Liu , Huanzhang Dou , Senqiao Yang , Sitong Wu , Hanbin Zhao , Jiaya Jia