About me
I am currently a CS PhD student in AIRS Lab from Stony Brook University, advised by Prof. Yi Liu.
My research interests include probabilistic generative models, geometric learning, and multimodal/graph learning.
I’m seeking an internship. Please feel free to send me the email for any potential opportunities.
Education
- B.Eng. Internet of Things Engineering; Jiangsu University of Science and Technology, Aug. 2017 - Jun. 2021
- M.Eng. Traffic Information Engineering and Control; Wuhan University of Technology, Aug. 2021 - Jun. 2024
- Ph.D. Computer Science; Stony Brook University, Aug. 2024 - TBD
News
- March 2026: Pass the RPE and become a PhD candidate!
- Jan. 2026: First author on a paper accepted by ICLR 2026 on efficient 3D molecular generation!
- Dec. 2025: Accepted the offer from Autodesk to work as an intern research scientist over the 2026 summer.
- Oct. 2025: We published Paper Finder, a tool designed to help researchers efficiently find publications from leading AI conferences, with support for multilingual semantic search on abstracts.
- Sept. 2025: IAGA is accepted by NeuRIPS 2025 SPIGM Workshop.
- July 2025: Preprint IAGA, a method for efficient 3D molecular generation via shortening generation trajectories.
- May 2025: Received IACS Young Writer’s Award, thank you IACS!
- May 2025: One paper accepted by ACL 2025 on LLM for molecule editing, congrats to my collaborators!
- May 2025: First author on a paper accepted by ICML 2025 on 3D GNN explanation (XAI)!
- April 2025: Give a lightening talk in IACS research day. Thanks for the opportunity from IACS!
First or Co-first Author Publications
* indicates equal contributors.
GAGA: Gaussianity-aware Gaussian Approximation for Efficient 3D Molecular Generation
Published in The Fourteenth International Conference on Learning Representations (ICLR 2026), 2026
We propose an analytical Gaussian approximation method to improve the efficiency of 3D molecular generation.
Recommended citation: Qu, J., Gao, W., Xu, R. & Liu, Y. GAGA: Gaussianity-aware Gaussian Approximation for Efficient 3D Molecular Generation. In Fourteenth International Conference on Learning Representations (ICLR).
Download Paper
RISE: Radius of Influence based Subgraph Extraction for 3D Molecular Graph Explanation
Published in The Forty-second International Conference on Machine Learning (ICML), 2025
We develop an new pipeline for 3D-GNN explanation.
Recommended citation: Qu, J., Gao, W., Zhang, J., Liu, X., Wei, H., Ling, H., & Liu, Y. RISE: Radius of Influence based Subgraph Extraction for 3D Molecular Graph Explanation. In Forty-second International Conference on Machine Learning.
Download Paper
Double domain guided real-time low-light image enhancement for ultra-high-definition transportation surveillance
Published in IEEE Transactions on Intelligent Transportation Systems, 2024
We firstly explicitly incorporate gradient-domain enhancement for low-light images
Recommended citation: Qu, J., Liu, R. W., Gao, Y., Guo, Y., Zhu, F., & Wang, F. Y. (2024). Double domain guided real-time low-light image enhancement for ultra-high-definition transportation surveillance. IEEE Transactions on Intelligent Transportation Systems, 25(8), 9550-9562.
Download Paper