AI Threat Alert indexes 4,319+ peer-reviewed and preprint papers on AI/ML security — covering adversarial attacks, model defenses, red-teaming benchmarks, surveys, and security tooling. Papers are sourced from arXiv, classified by type and by relevance to real-world threats, and cross-referenced with the CVEs and incidents they relate to.
Showing 61–80 of 609 papers
Clear filtersChristoph Bühler, Matteo Biagiola, Luca Di Grazia +1 more
AI agents built on large language models (LLMs) run shell commands, read and write files, and reach the network, typically with their user's...
Tural Hagverdiyev
A tool-using model can follow a malicious instruction even when its credentials are valid. We study whether task-scoped authorization contains the...
Haocheng Yang, Yuchao Zhang, Licheng Pan +8 more
Rubric-based reinforcement learning (RL) provides interpretable rewards for aligning large language models (LLMs) by evaluating responses against...
Maria Carmen Jica, Ali Satvaty, Suzan Verberne +1 more
Multimodal Retrieval-Augmented Generation (MRAG) has emerged as a reliable and cost-effective technique of grounding the generative capabilities of...
Tian Dong, Zixuan Ma, Haodong Zhao +3 more
LLM agents use skills to improve performance on specialized tasks. To complete a user request, an agent may invoke several skills in sequence,...
Kaiyang Li, Jiahao Chen, Yuwen Pu +5 more
Safety-aligned Large Language Models remain vulnerable to fine-tuning on small sets of harmful or benign-looking samples. However, prior studies...
Marcelo Fernandez
A correctly governed LLM agent can reach a state in which neither continuing execution nor automatically halting is admissible: the system has...
Birk Torpmann-Hagen, Finn Schwall, Leon Moonen
Autonomous large language model (LLM) agents increasingly interact in network environments where adversarial content can propagate between agents....
Jieke Shi, Yuchen Chen, Junda He +2 more
Repository instruction files guide coding agents, but also expose them to prompt injection. Malicious rules can request credential access or data...
Shouli Wang, Yanfeng Jia, Zhihao Ou +6 more
During reinforcement learning with verifiable rewards (RLVR), large language models (LLMs) can exploit loopholes in their environments to obtain high...
Shouli Wang, Yanfeng Jia, Zhihao Ou +6 more
During reinforcement learning with verifiable rewards (RLVR), large language models (LLMs) can exploit loopholes in their environments to obtain high...
Soohan Lim, Hyundong Jin, Yo-Sub Han
Code watermarking supports provenance tracking for code generated by LLMs. Modifying token selection to embed watermarks as an LLM generates code can...
Wenxin Wu, Lingyong Yan, Lei Sha +2 more
LLM agents increasingly rely on reusable Skills for complex, multi-step tasks, creating a critical supply-chain attack surface where poisoned Skill...
Yan Wang, Zhihao Zhang, Ke Chen +5 more
LLM agents increasingly rely on installable skills, which are packages of instructions, code, and resources that equip them with task-specific...
Yaohua Liu, Yifan Guo, Jiaxin Gao
A key bottleneck in adversarial transfer is a trajectory-level geometric disconnect: ambient gradients often drift away from the intrinsic data...
Yunjia Zheng, Bintang Dwi Marthen, Zachary Pan +5 more
Coding agents have become real users of high-performance computing (HPC) systems, yet today's HPC abstractions, interfaces, and policies remain...
Qiuyu Ren, Sudipta Paria, Aritra Dasgupta +1 more
Large language models (LLMs) incur substantial storage, memory-bandwidth and energy costs, motivating compact weight representations. Existing...
Shuming Liu, Zhifang Zhang, Suqin Yuan +3 more
Vision-language models (VLMs) exhibit strong multimodal capabilities but remain vulnerable to backdoors implanted through poisoned fine-tuning data....
Zewen Sun, Tongyang Zhao, Liyao Xiang +3 more
Semantic watermarking improves robustness against watermark removal attacks by embedding detectable signals into sentence-level representations....
Haoran Ou, Gelei Deng, Xuanye Zhang +3 more
As LLM-based agents perform increasingly complex tasks, Agent Skills have emerged as a flexible mechanism for extending their capabilities. An Agent...
AI security research studies how AI and machine-learning systems can be attacked and defended — covering adversarial examples, prompt injection, model poisoning, training-data extraction, and the mitigations against them. AI Threat Alert curates this research from academic sources so security teams can track the threats behind emerging AI risks.
AI Threat Alert indexes 4,319+ papers on AI/ML security, classified across attack, defense, benchmark, survey, and tool categories and updated continuously.
Papers are sourced from arXiv, then classified by type and by relevance to real-world AI/ML threats, and cross-referenced with the CVEs and incidents they relate to.
Coverage spans adversarial attacks, model and system defenses, red-teaming benchmarks, literature surveys, and security tooling for LLMs, ML libraries, AI agents, and inference pipelines.
Every paper is filtered for AI security relevance and linked to the vulnerabilities, vendors, and incidents it relates to, so the research connects directly to operational threat intelligence.
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