Our research lab is dedicated to advancing visual understanding through cutting-edge AI and machine learning techniques. Our mission is to explore and develop innovative solutions in visual language models (VLMs), 3D vision, and more, to push the boundaries of how machines perceive and interpret the world.
- [Under-review] [YesBut-v2] When ‘YES’ Meets ‘BUT’: Can AI Comprehend Contradictory Humor Through Comparative Reasoning?
- [Under-review] When Words Outperform Vision: VLMs Can Self-Improve Via Text-Only Training For Human-Centered Decision Making
- [NeurIPS 2024 Oral] [YesBut] Cracking the Code of Juxtaposition: Can AI Models Understand the Humorous Contradictions?
- [EMNLP 2024] VIVA: A Benchmark for Vision-Grounded Decision-Making with Human Values
- [INLG 2024 Oral] AMERICANO: Argument Generation with Discourse-driven Decomposition and Multi-agent Interaction
Debate-to-Write: A Persona-Driven Multi-Agent Framework for Diverse Argument Generation International Conference on Computational Linguistics (COLING), 2025
- [Under-review] Segment-then-Splat: A Unified Approach for 3D Open-Vocabulary Segmentation based on Gaussian Splatting
- [Under-review] CAUSAL3D: A Comprehensive Benchmark for Causal Learning from Visual Data
- [CVPR 2025] BARD-GS: Blur-Aware Reconstruction of Dynamic Scenes via Gaussian Splatting
- [ACM MM 2024] View-consistent Object Removal in Radiance Fields
- [CVPR 2023] NeRFInvertor: High Fidelity NeRF-GAN Inversion for Single-shot Real Image Animation
Explore these projects and more to learn how we're shaping the future of visual AI!