NeurIPS 2025: First author (equal contribution) of 'Imagine360: Immersive 360 Video Generation from Perspective Anchor'.
SIGGRAPH 2025 (Conference Track): Equal contribution on 'LayerPano3D: Layered 3D Panorama for Hyper-Immersive Scene Generation'.
T-PAMI 2023: First author of 'Temporal Perceiver: A General Architecture for Arbitrary Boundary Detection'.
NeurIPS 2022: First author of 'PointTAD: Multi-Label Temporal Action Detection with Learnable Query Points'.
ICCV 2021: Equal contribution on 'Relaxed Transformer Decoders for Direct Action Proposal Generation'—the first transformer-based framework for temporal action proposal generation.
ICCV 2025: Co-author of 'GenDoP: Auto-regressive Camera Trajectory Generation as a Director of Photography'.
ICLR 2025: Co-author of 'IDArb: Intrinsic Decomposition for Arbitrary Number of Input Views and Illuminations'.
NeurIPS D&B Track 2024: Contributed to 'HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation'.
CVPR 2024: Co-author of 'Dual DETRs for Multi-Label Temporal Action Detection'.
Regular reviewer for CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR; reviewer for SIGGRAPH 2025; journal reviewer for IJCV.
Teaching Assistant for IERG4998 and IERG4999 at CUHK.