GenAI Security Literature Review
A comprehensive, community-driven, auto-updating database of GenAI and LLM security research, standards, tools, and resources.
Recent Additions
IoT-Aided Self-Adaptive Routing for Flying Ad Hoc Network Using Gauss-Markov Mobility Model
Oroos Arshi, Tarandeep Kaur Bhatia, Inam Ullah Khan + 2 more
Introduction: Unmanned aerial vehicles (UAVs) are low-cost, easy to deploy, and can be integrated with IoT to solve human-related problems. They have applications in military and civil sectors, such as forest fire detection, sports monitoring, agriculture, and border surveillance. UAV-based applications can be signal or multi-system.
MetaBackdoor: Exploiting Positional Encoding as a Backdoor Attack Surface in LLMs
Rui Wen, Mark Russinovich, Andrew Paverd + 2 more
Backdoor attacks pose a serious security threat to large language models (LLMs), which are increasingly deployed as general-purpose assistants in safety- and privacy-critical applications. Existing LLM backdoors rely primarily on content-based triggers, requiring explicit modification of the input text. In this work, we show that this assumption is unnecessary and limiting. We introduce MetaBackdoor, a new class of backdoor attacks that exploits positional information as the trigger, without mod
Talk is (Not) Cheap: A Taxonomy and Benchmark Coverage Audit for LLM Attacks
Karthik Raghu Iyer, Yazdan Jamshidi, Nicholas Bray + 1 more
We introduce a reusable framework for auditing whether LLM attack benchmarks collectively cover the threat surface: a 4$\times$6 Target $\times$ Technique matrix grounded in STRIDE, constructed from a 507-leaf taxonomy -- 401 data-populated and 106 threat-model-derived leaves -- of inference-time attacks extracted from 932 arXiv security studies (2023--2026). The matrix enables benchmark-external validation -- auditing collective coverage rather than individual benchmark consistency. Applying it
Model-Agnostic Lifelong LLM Safety via Externalized Attack-Defense Co-Evolution
Xiaozhe Zhang, Chaozhuo Li, Hui Liu + 4 more
Large language models remain vulnerable to adversarial prompts that elicit harmful outputs. Existing safety paradigms typically couple red-teaming and post-training in a closed, policy-centric loop, causing attack discovery to suffer from rapid saturation and limiting the exposure of novel failure modes, while leaving defenses inefficient, rigid, and difficult to transfer across victim models. To this end, we propose EvoSafety, an LLM safety framework built around persistent, inspectable, and re
Quantifying LLM Safety Degradation Under Repeated Attacks Using Survival Analysis
Zvi Topol
Large language models (LLMs) are increasingly deployed in a wide range of applications, yet remain vulnerable to adversarial jailbreak attacks that circumvent their safety guardrails. Existing evaluation frameworks typically report binary success/failure metrics, failing to capture the temporal dynamics of how attacks succeed under persistent adversarial pressure. This preliminary work proposes a novel evaluation framework that applies survival analysis techniques to characterize LLM jailbreak v
Most Cited
ReAct: Synergizing Reasoning and Acting in Language Models
Shunyu Yao, Jeffrey Zhao, Dian Yu + 4 more — ICLR 2023
Foundational work on the ReAct paradigm for LLM agents that interleave reasoning and tool-use actions, enabling complex task completion with security implications.
Toolformer: Language Models Can Teach Themselves to Use Tools
Timo Schick, Jane Dwivedi-Yu, Roberto Dessi + 6 more — NeurIPS 2023
Demonstrates how LLMs can learn to use external tools (APIs, search engines, calculators) through self-supervised learning, foundational for understanding tool-use security.
Extracting Training Data from Large Language Models
Nicholas Carlini, Florian Tramer, Eric Wallace + 9 more — USENIX Security 2021
Demonstrates that large language models memorize and can be prompted to emit verbatim training data, including PII, revealing significant privacy risks.
Constitutional AI: Harmlessness from AI Feedback
Yuntao Bai, Saurav Kadavath, Sandipan Kundu + 44 more — arXiv preprint
Introduces Constitutional AI (CAI), a method for training AI systems to be harmless using a set of principles (a constitution) and AI-generated feedback, reducing reliance on human red teamers.
Universal and Transferable Adversarial Attacks on Aligned Language Models
Andy Zou, Zifan Wang, Nicholas Carlini + 3 more — arXiv preprint
Proposes an automated method (GCG) to generate adversarial suffixes that cause aligned LLMs to produce harmful content, with attacks transferring across models including ChatGPT and Claude.