collaborators

9 papers

cs.CR2026

Evading Chain-of-Thought Monitoring Through Model Poisoning

Giorgio Severi, Shujaat Mirza, Blake Bullwinkel +1

Chain-of-thought (CoT) monitoring is an increasingly important component of AI safety stacks but relies on the assumption that a model's reasoning trace is informative about its ac…

cs.CL2026

Learning to Attack and Defend: Adaptive Red Teaming of Language Models via GRPO

Blake Bullwinkel, Eugenia Kim, Amanda Minnich +1

AI red teaming must continually adapt to evolving attackers and defenders. Reinforcement learning offers a promising approach to discovering novel attacks, and co-training methods…

cs.HC2026

Beyond the Single Turn: Reframing Refusals as Dynamic Experiences Embedded in the Context of Mental Health Support Interactions with LLMs

Ningjing Tang, Alice Qian, Qiaosi Wang +6

Content Warning: This paper contains participant quotes and discussions related to mental health challenges, emotional distress, and suicidal ideation. Large language models (LLMs)…

cs.LG2026

GRP-Obliteration: Unaligning LLMs With a Single Unlabeled Prompt

Mark Russinovich, Yanan Cai, Keegan Hines +3

Safety alignment is only as robust as its weakest failure mode. Despite extensive work on safety post-training, it has been shown that models can be readily unaligned through post-…

cs.CR2026

The Trigger in the Haystack: Extracting and Reconstructing LLM Backdoor Triggers

Blake Bullwinkel, Giorgio Severi, Keegan Hines +3

Detecting whether a model has been poisoned is a longstanding problem in AI security. In this work, we present a practical scanner for identifying sleeper agent-style backdoors in…

cs.CR2025

A Systematization of Security Vulnerabilities in Computer Use Agents

Daniel Jones, Giorgio Severi, Martin Pouliot +7

Computer Use Agents (CUAs), autonomous systems that interact with software interfaces via browsers or virtual machines, are rapidly being deployed in consumer and enterprise enviro…