news

Feb 10, 2026 Our paper GRP-Obliteration: Unaligning LLMs With a Single Unlabeled Prompt is now available on ArXiv
Feb 5, 2026 Our paper The Trigger in the Haystack: Extracting and Reconstructing LLM Backdoor Triggers is now available on ArXiv
Jul 8, 2025 Our paper A Systematization of Security Vulnerabilities in Computer Use Agents is now available on ArXiv
Jun 4, 2025 Our paper A Representation Engineering Perspective on the Effectiveness of Multi-Turn Jailbreaks is now available on ArXiv
Jun 1, 2025 Our paper Weathering the CUA Storm: Mapping Security Threats in the Rapid Rise of Computer Use Agents was accepted at the ICML 2025 Workshop on Computer Use Agents!
Jan 14, 2025 Our paper Lessons From Red Teaming 100 Generative AI Products is now available on ArXiv
Jan 1, 2025 Our paper Model-Agnostic Clean-Label Backdoor Mitigation in Cybersecurity Environments was accepted at MILCOM 2025!
Jun 4, 2024 Our new paper Phantom: General Trigger Attacks on Retrieval Augmented Language Generation is now available on ArXiv
Jan 15, 2024 Our paper Chameleon: Increasing Label-Only Membership Leakage with Adaptive Poisoning was accepeted at ICLR 2024!
Dec 7, 2023 I am presenting our paper Poisoning Network Flow Classifiers at the Annual Computer Security Applications Conference (ACSAC)
Jun 1, 2022 I am returning to Microsoft as an intern in the AI Red Team this summer.
Jun 1, 2021 I will join the Microsoft Azure Trustworthy Machine Learning team as an intern this summer.
May 1, 2019 I will join FireEye Data Science as an intern this summer.