Sep 2025: Preprint 'GHOST: Hallucination-Inducing Image Generation for Multimodal LLMs' posted on arXiv
Aug 2025: Paper 'Gaming Tool Preferences in Agentic LLMs' accepted to EMNLP 2025 (Main Track)
Apr 2025: Preprint 'Chain-of-Defensive-Thought: Structured Reasoning Elicits Robustness in Large Language Models against Reference Corruption' posted on arXiv
Mar 2025: Preprint 'Seeing What’s Not There: Spurious Correlation in Multimodal LLMs' posted on arXiv
Dec 2024: Co-organizing the 'Spurious Correlation and Shortcut Learning Workshop' at ICLR 2025
Jun 2024: Paper 'Decompose-and-Compose: A Compositional Approach to Mitigating Spurious Correlation' presented at CVPR 2024