Publications
Preprints & under submission
No Attacker Needed: Unintentional Cross-User Contamination in Shared-State LLM Agents
Under submission, 2026
Learning to Route LLMs from Bandit Feedback: One Policy, Many Trade-offs
Under submission, 2025
StealthRank: LLM Ranking Manipulation via Stealthy Prompt Optimization
Under submission, 2025
Conference papers
A Personalized Conversational Benchmark: Towards Simulating Personalized Conversations
NeurIPS MTI-LLM Workshop, 2025 Spotlight
AD-AGENT: A Multi-agent Framework for End-to-end Anomaly Detection
IJCNLP-AACL Findings, 2025
DPU: Dynamic Prototype Updating for Multimodal Out-of-Distribution Detection
CVPR, 2025 Highlight
PyOD 2: A Python Library for Outlier Detection with LLM-powered Model Selection
The Web Conference (Demo Track), 2025