Chonghan Qin
秦崇瀚
Ph.D. Student [AT] CUHK MMLab
qinchonghanzuibang [AT] gmail [DOT] com
Bio
I am a Ph.D. student at MMLab, The Chinese University of Hong Kong, advised by Prof. Xiangyu Yue and Prof. Dahua Lin. Previously, I received my B.Eng. from The University of Hong Kong. During my undergraduate studies, I had the privilege of working with Dr. Xiachong Feng and Prof. Lingpeng Kong at HKUNLP, Dr. Lijun Wu at Shanghai Artificial Intelligence Laboratory, and Dr. Weijie Chen at Fudan University.
I study (Multimodal) Large Language Models and Embodied AI.
News
Jul 2026 ImplicitMemBench received an ACL 2026 Best Resource Award, one of only 4 papers selected (~0.17% of accepted papers), and was also selected as an SAC Highlight (~2.9%).
Jun 2026 As I wrap up undergrad, I want to thank the advisors, mentors, friends, family, and everyone else who made these years what they were. I couldn’t have done it without you. Read acknowledgements (PDF)
Jun 2026 SciGenBench and ImgCoder was accepted to ECCV 2026. Thanks again to all collaborators!
May 2026 ImplicitMemBench has been selected for an Oral Presentation at ACL 2026 Main Conference in the Resources and Evaluation track. See you all in San Diego.
May 2026 We introduced The Granularity Axis, an attempt to find the hidden micro-to-macro slider behind how LLMs reason as individuals, institutions, and nations.
Apr 2026 ImplicitMemBench was accepted to ACL 2026 Main Conference, and SAVOIR was accepted to ACL 2026 Findings. Thanks again to all collaborators!
Jan 2026 We released MMFineReason, taking a data-centric route to improve multimodal reasoning: our 4B VLM reaches 30B-level performance. The corresponding reasoning dataset MMFineReason-1.8M reached #2 on HuggingFace Datasets Trending with 25k+ downloads shortly after release.
Jan 2026 We introduced SciGenBench and ImgCoder, aiming to make scientific image synthesis focus more on structure, precision, and actual reasoning.
Nov 2025 We placed #4 in DCVLR, a vision-language reasoning competition at the NeurIPS 2025 Workshop.
Publications
Most recent publications on Google Scholar.
† indicates equal contribution.
The Granularity Axis: A Micro-to-Macro Latent Direction for Social Roles in Language Models
Chonghan Qin, Xiachong Feng, Ziyun Song, Xiaocheng Feng, Jing Xiong, Lingpeng Kong
ImplicitMemBench: Measuring Unconscious Behavioral Adaptation in Large Language Models
Chonghan Qin, Xiachong Feng, Weitao Ma, Xiaocheng Feng, Lingpeng Kong
SAVOIR: Learning Social Savoir-Faire via Shapley-based Reward Attribution
Xiachong Feng, Yi Jiang, Xiaocheng Feng, Deyi Yin, Libo Qin, Yangfan Ye, Lei Huang, Weitao Ma, Yuxuan Gu, Chonghan Qin, Bing Qin, Lingpeng Kong
Scientific Image Synthesis: Benchmarking, Methodologies, and Downstream Utility
Honglin Lin†, Chonghan Qin†, Zheng Liu, Qizhi Pei, Yu Li, Zhanping Zhong, Xin Gao, Yanfeng Wang, Conghui He, Lijun Wu
MMFineReason: Closing the Multimodal Reasoning Gap via Open Data-Centric Methods
Honglin Lin†, Zheng Liu†, Yun Zhu†, Chonghan Qin, Juekai Lin, Xiaoran Shang, Conghui He, Wentao Zhang, Lijun Wu
ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch
Zheng Liu†, Honglin Lin†, Chonghan Qin, Xiaoyang Wang, Xin Gao, Yu Li, Mengzhang Cai, Yun Zhu, Zhanping Zhong, Qizhi Pei, Zhuoshi Pan, Xiaoran Shang, Bin Cui, Conghui He, Wentao Zhang, Lijun Wu
The Granularity Axis: A Micro-to-Macro Latent Direction for Social Roles in Language Models
Chonghan Qin, Xiachong Feng, Ziyun Song, Xiaocheng Feng, Jing Xiong, Lingpeng Kong
ImplicitMemBench: Measuring Unconscious Behavioral Adaptation in Large Language Models
Chonghan Qin, Xiachong Feng, Weitao Ma, Xiaocheng Feng, Lingpeng Kong
SAVOIR: Learning Social Savoir-Faire via Shapley-based Reward Attribution
Xiachong Feng, Yi Jiang, Xiaocheng Feng, Deyi Yin, Libo Qin, Yangfan Ye, Lei Huang, Weitao Ma, Yuxuan Gu, Chonghan Qin, Bing Qin, Lingpeng Kong
Scientific Image Synthesis: Benchmarking, Methodologies, and Downstream Utility
Honglin Lin†, Chonghan Qin†, Zheng Liu, Qizhi Pei, Yu Li, Zhanping Zhong, Xin Gao, Yanfeng Wang, Conghui He, Lijun Wu
MMFineReason: Closing the Multimodal Reasoning Gap via Open Data-Centric Methods
Honglin Lin†, Zheng Liu†, Yun Zhu†, Chonghan Qin, Juekai Lin, Xiaoran Shang, Conghui He, Wentao Zhang, Lijun Wu
ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch
Zheng Liu†, Honglin Lin†, Chonghan Qin, Xiaoyang Wang, Xin Gao, Yu Li, Mengzhang Cai, Yun Zhu, Zhanping Zhong, Qizhi Pei, Zhuoshi Pan, Xiaoran Shang, Bin Cui, Conghui He, Wentao Zhang, Lijun Wu
Service
Teaching:
The University of Hong Kong
2025-2026 AILT1001: Artificial Intelligence Literacy I, Lead Student Teaching Assistant
2024 SCDS1001: Artificial Intelligence Literacy I, Student Teaching Assistant
Academic Advising:
The University of Hong Kong
2025-2026 Class of 2029, School of Computing and Data Science, around 20 students
2024 Class of 2028, Department of Computer Science, around 15 students
Journal Reviewer:
- IEEE Transactions on Big Data: 2025
Vitæ
-
The Chinese University of Hong Kong 2026 Aug - PresentPh.D. Student
Multimedia Laboratory, Information Engineering -
Shanghai Artificial Intelligence Laboratory 2025 June - 2026 MayResearch Intern
OpenDataLab -
The University of Hong Kong 2024 Oct. - 2026 MayStudent Research Assistant
HKUNLP Group -
Fudan University 2024 Aug - 2024 Dec.Research Intern
Intelligent Medicine Institute -
E Fund Management (HK) Co., Ltd 2024 June - 2024 AugQuantitative Research Intern
Equity, Fixed Income -
University of California, Berkeley 2023 Su, 2024 FaExchange Student
Electronic Engineering and Computer Science -
Tishman Speyer 2023 Dec - 2024 JanIntern
Acquisition -
The University of Hong Kong 2022 Sept. - 2026 MayB.Eng. Student
Computer Science