The PolyU COMP - HKUST (GZ) INFH Research Student Conference was successfully held on 19 June 2025 at The Hong Kong University of Science and Technology (Guangzhou) (HKUST (GZ)). Co-hosted by the Department of Computing (COMP) and the Information Hub (INFH), HKUST (GZ), the event brought together nearly 400 research students and faculty members, fostering a dynamic platform for academic exchange, research showcase, and cross-institutional collaboration.
The conference commenced with a Welcome Speech by Prof. Lionel M. NI, President of HKUST (GZ), followed by Opening Speeches from Prof. Lei CHEN, Dean of the INFH, and Prof. Qing LI, Chair Professor of Data Science and Head of COMP. Both speakers underscored the importance of nurturing young researchers and promoting interdisciplinary collaboration.
Technical Program Committee (TPC) Remarks were delivered by Prof. Lei YANG, Associate Professor of COMP, who presented an overview of paper submissions, the review process, and final decisions, while also highlighting the transformative impact of AI on research across diverse fields.
A key highlight of the conference was the Keynote Speech by Prof. James KWOK, Professor of the Department of Computer Science and Engineering at HKUST, who shared insights on “Achieving Multiple Goals with Ease”, inspiring attendees with his expertise and vision.
The conference featured eight oral presentation sessions and a poster session, covering a wide range of research areas including:
- Embodied Intelligence and Robotics
- Cyber Security and Privacy
- Networking and Mobile Computing
- Data Science and Information Retrieval
- Theory, Optimization, and Software Engineering
- Vision, Graphics, and Human-Computer Interaction
- Natural Language Processing
- AI for Science
A total of 79 papers and 89 posters were selected by the TPC for presentation, offering students a valuable opportunity to present their work, receive feedback, and engage in scholarly dialogue.
The day concluded with a Dinner Banquet and Award Presentation at the Sheraton Guangzhou Nansha Hotel. Awards were presented to eight students for Best Oral Presentation, selected by expert panels composed of faculty members from COMP and INFH, and ten students for Best Poster, voted by all participants. (Awardee list in alphabetical order of surnames.):
Best Oral Presentation Award
University |
Awardees |
Paper Title |
PolyU |
SHEN Leming |
AutoIOT: LLM-Driven Automated Natural Language Programming for AIoT Applications |
XIA Heming |
TokenSkip: Controllable Chain-of-Thought Compression in LLMs |
|
XING Shiji |
Allocating Resources with Imperfect Information |
|
ZHANG Xin |
GME: Improving Universal Multimodal Retrieval by Multimodal LLMs |
|
HKUST(GZ) |
MA Mulei |
FlocOff: Data Heterogeneity Resilient Federated Learning with Communication-Efficient Edge Offloading |
WANG Yingna |
Blossoms Across Time: AI-Assisted Cultural Dialogue Through Diverse Artistic Expressions in VR Intangible Cultural Heritage Experience |
|
YANG Xun |
Omni-Perception: Omnidirectional Collision Avoidance for Legged Locomotion in Dynamic Environment |
|
ZHENG Jingyi |
TH-Bench: Benchmarking Humanization-based Evading Attacks Against Machine- Generated Text Detectors |
Best Poster Presentation Award
University |
Awardees |
Poster Title |
PolyU |
DONG Yiying |
Triplet-Based Referring Paradigm for Unified Visual Decoding |
HUANG Xinyu |
Towards Privacy-Preserving and Personalized Smart Home Serving through Tailored Small Language Models |
|
JIANG Jinan |
ConsCS: Effective and Efficient Verification of Circom Circuits |
|
LIU Xun |
SnarkFold: Efficient Proof Aggregation from Incremental Verification and Applications |
|
LYU Xinqi |
PLA: Prompt Learning Attack against Text-to-Image Generative Models |
|
WANG Zhicheng |
From FSD to FSC: Enabling Full Smart-Communication in Autonomous Vehicles Through Full Self-Driving Models |
|
WU Yuhui |
InsViE-1M: Effective Instruction-based Video Editing with Elaborate Dataset Construction |
|
XIAO Chaodong |
Spatial-Mamba: Effective Visual State Space Models via Structure-Aware State Fusion |
|
YANG Chang |
Nondeterministic Polynomial-time Problem Challenge: An Ever-Scaling Reasoning Benchmark for LLMs |
|
ZHOU Yi |
Computational Protein Science in the Era of Large Language Models (LLMs) |
Closing Speeches were delivered by Prof. Lei CHEN and Prof. Qing LI, who expressed appreciation for the enthusiastic participation and reaffirmed the commitment to continued collaboration between PolyU and HKUST (GZ).
The PolyU COMP - HKUST (GZ) INFH Research Student Conference was a resounding success, further strengthening the partnership ties between the two institutions and empowering the next generation of researchers through meaningful academic exchange.