6 papers
Rapid Poison: Practical Poisoning Attacks Against the Rapid Response Framework
David Huang, Jaewon Chang, Avidan Shah +2
The Rapid Response (RR) framework, deployed in production systems, including Anthropic's ASL-3 safeguards, continuously improves jailbreak-detection classifiers. When new jailbreak…
Soft Instruction De-escalation Defense
Nils Philipp Walter, Chawin Sitawarin, Jamie Hayes +2
Large Language Models (LLMs) are increasingly deployed in agentic systems that interact with an external environment; this makes them susceptible to prompt injections when dealing…
The Llama 4 Herd: Architecture, Training, Evaluation, and Deployment Notes
Redacted by arXiv
This document consolidates publicly reported technical details about Metas Llama 4 model family. It summarizes (i) released variants (Scout and Maverick) and the broader herd conte…
Extracting alignment data in open models
Federico Barbero, Xiangming Gu, Christopher A. Choquette-Choo +6
In this work, we show that it is possible to extract significant amounts of alignment training data from a post-trained model -- useful to steer the model to improve certain capabi…
The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections
Milad Nasr, Nicholas Carlini, Chawin Sitawarin +11
How should we evaluate the robustness of language model defenses? Current defenses against jailbreaks and prompt injections (which aim to prevent an attacker from eliciting harmful…
JailbreaksOverTime: Detecting Jailbreak Attacks Under Distribution Shift
Julien Piet, Xiao Huang, Dennis Jacob +7
Safety and security remain critical concerns in AI deployment. Despite safety training through reinforcement learning with human feedback (RLHF) [ 32], language models remain vulne…