2,560+ academic papers on AI security, attacks, and defenses
Showing 241–260 of 1,936 papers
Clear filtersEuntae Kim, Soomin Han, Buru Chang
Large language models (LLMs) are increasingly used as co-authors in collaborative writing, where users begin with rough drafts and rely on LLMs to...
Kun Wang, Cheng Qian, Miao Yu +6 more
Multimodal Large Language Models (MLLMs) have achieved remarkable success in cross-modal understanding and generation, yet their deployment is...
MinJae Jung, YongTaek Lim, Chaeyun Kim +3 more
While Large Language Models (LLMs) are widely used, they remain susceptible to jailbreak prompts that can elicit harmful or inappropriate responses....
Hugo Lyons Keenan, Christopher Leckie, Sarah Erfani
We can often verify the correctness of neural network outputs using ground truth labels, but we cannot reliably determine whether the output was...
Jiacheng Liang, Yao Ma, Tharindu Kumarage +5 more
Reinforcement Learning from Human Feedback (RLHF) is central to aligning Large Language Models (LLMs), yet it introduces a critical vulnerability: an...
Hanrui Luo, Shreyank N Gowda
Detecting jailbreak behaviour in large language models remains challenging, particularly when strongly aligned models produce harmful outputs only...
Ahson Saiyed, Sabrina Sadiekh, Chirag Agarwal
Large Language Models (LLMs) remain vulnerable to optimization-based jailbreak attacks that exploit internal gradient structure. While Sparse...
Ruixuan Liu, David Evans, Li Xiong
Indistinguishability properties such as differential privacy bounds or low empirically measured membership inference are widely treated as proxies to...
Md Rysul Kabir, Zoran Tiganj
Open-weight language models can be rendered unsafe through several distinct interventions, but the resulting models may differ substantially in...
Thamilvendhan Munirathinam
Current open-source prompt-injection detectors converge on two architectural choices: regular-expression pattern matching and fine-tuned transformer...
Sina Abdollahi, Mohammad M Maheri, Javad Forough +5 more
Large Language Model (LLM) agents provide powerful automation capabilities, but they also create a substantially broader attack surface than...
Yanzhen Lu, Muchen Jiang, Zhicheng Qian +1 more
Prompt-injected memory can improve reasoning without updating model weights, but it also creates a control problem: retrieved content helps only when...
Ziyang Liu
Hosted-LLM providers have a silent-substitution incentive: advertise a stronger model while serving cheaper replies. Probe-after-return schemes such...
Wentao Zhang, Yan Zhuang, ZhuHang Zheng +3 more
Existing jamming attacks on Retrieval-Augmented Generation (RAG) systems typically induce explicit refusals or denial-of-service behaviors, which are...
Sua Lee, Sanghee Park, Jinbae Im
Multimodal Large Language Models (MLLMs) have been increasingly used as automatic evaluators-a paradigm known as MLLM-as-a-Judge. However, their...
Jin Zhao, Marta Knežević, Tanja Käser
Large Language Models (LLMs) are increasingly used in education, yet their default helpfulness often conflicts with pedagogical principles. Prior...
Hu Wei
AI agent systems increasingly rely on reusable non-LLM engineering infrastructure that packages tool mediation, context handling, delegation, safety...
Dongcheng Zhang, Yiqing Jiang
Existing AI agent safety benchmarks focus on generic criminal harm (cybercrime, harassment, weapon synthesis), leaving a systematic blind spot for a...
Xingyu Liu, Zengqin Huang, Xiang Gao +1 more
Fuzz testing of software libraries relies on fuzz drivers to invoke library APIs. Traditionally, these drivers are written manually by developers - a...
Parteek Jamwal, Minghao Shao, Boyuan Chen +15 more
Large Language Models (LLMs) have demonstrated remarkable capabilities across various cybersecurity tasks, including vulnerability classification,...
Get breaking CVE alerts, compliance reports (ISO 42001, EU AI Act), and CISO risk assessments for your AI/ML stack.
Start 14-Day Free Trial
AI Threat Alert