ICLR 2025 Spotlight: Interleaved Scene Graph for Interleaved Text-and-Image Generation Assessment
ICLR 2025: GUI-World: A Dataset for GUI-oriented Multimodal LLM-based Agents
NeurIPS 2024: The Best of Both Worlds: Toward an Honest and Helpful Large Language Model
ICML 2024 Oral: MLLM-as-a-Judge: Assessing Multimodal LLM-as-a-Judge with Vision-Language Benchmark
NAACL 2024 Findings: LLM-as-a-Coauthor: The Challenges of Detecting LLM-Human Mixcase
Tech Report: Reinforced Visual Perception with Tools
Preprint: Are We on the Right Way for Assessing Document Retrieval-Augmented Generation?
Preprint: code_transformed: The Influence of Large Language Models on Code
Tech Report: Optimizing Length Compression in Large Reasoning Models
Preprint: Seeking and Updating with Live Visual Knowledge
ACM MM 2025 Dataset Track: MultiRef: Controllable Image Generation with Multiple Visual References
EMNLP 2025 Findings: Wait, We Don’t Need to “Wait”! Removing Thinking Tokens Improves Reasoning Efficiency
Research Experience
Currently working with Professors Tianyi Zhou and Sarah Wiegreffe at the University of Maryland; previously a visiting student at the University of Washington, collaborating with Professor Ranjay Krishna and Senior PhD Jieyu Zhang; leading ONE Lab and working with Professor Yao Wan.
Education
Pursuing a PhD at the University of Maryland, College Park, under the supervision of Professors Tianyi Zhou and Sarah Wiegreffe; previously a visiting student at the University of Washington, working with Professors Ranjay Krishna and Jieyu Zhang; also collaborated with Lichao Sun at Lehigh University and Xiangliang Zhang at the University of Notre Dame.
Background
Research interests: Latest advancements in generative models, especially multimodal, LLM-based agents, and frontier trustworthy problems. Currently a first-year PhD student at the University of Maryland, College Park.
Miscellany
Looking for summer intern opportunities in 2026; interested in connecting with students for research collaboration, mentorship, or just making friends; particularly encouraging junior students or students from underrepresented groups to reach out; feel free to schedule a meeting via this link.