Jin Wang

CS PhD Student at HKU

The University of Hong Kong

I am currently a 3rd-year PhD student in Computer Science at The University of Hong Kong (HKU), supervised by Prof. Ping Luo. My research focuses on multimodal foundation models, especially unified systems that connect visual understanding, generation, and evaluation. My earlier work explored deepfake detection and AI interpretability. I have been a research intern at Hunyuan (Tencent Project Up), Huawei Noah’s Ark Lab, and Megvii. Before joining HKU, I obtained my master’s degree from the University of Chinese Academy of Sciences under Prof. Chao Li, and received my bachelor’s degree from Dalian University of Technology (DLUT).

Education

PhD in Computer Science

The University of Hong Kong

October 2023 – July 2027 Hong Kong

MEng in Electronic Engineering

University of Chinese Academy of Sciences

September 2020 – June 2023 Beijing

BEng in Digital Media Technology

Dalian University of Technology

September 2016 – June 2020 Dalian, Liaoning

Publications

Unified Multimodal Understanding & Generation

FUDOKI: Discrete Flow-based Unified Understanding and Generation via Kinetic-Optimal Velocities

Jin Wang^* , Yao Lai^* , Aoxue Li , Shifeng Zhang , Jiacheng Sun , Ning Kang , Chengyue Wu , Zhenguo Li , Ping Luo * Equal contribution

In NeurIPS 2025 (Spotlight)

Multimodal Generation

TAGRPO: Boosting GRPO on Image-to-Video Generation with Direct Trajectory Alignment

Jin Wang^* , Jianxiang Lu^* , Guangzheng Xu , Comi Chen , Haoyu Yang , Linqing Wang , Peng Chen , Mingtao Chen , Zhichao Hu , Longhuang Wu , Shuai Shao , Qinglin Lu , Ping Luo * Equal contribution

To appear in ICML 2026

Rotate Your Character: Revisiting Video Diffusion Models for High-Quality 3D Character Generation

Jin Wang^* , Jianxiang Lu^* , Comi Chen , Guangzheng Xu , Haoyu Yang , Peng Chen , Na Zhang , Yifan Xu , Longhuang Wu , Shuai Shao , Qinglin Lu , Ping Luo * Equal contribution

Under Review

Megactor-sigma: Unlocking flexible mixed-modal control in portrait animation with diffusion transformer

Shurong Yang^* , Huadong Li^* , Juhao Wu^* , Minhao Jing^* , Linze Li , Renhe Ji , Jiajun Liang , Haoqiang Fan , Jin Wang * Equal contribution

In AAAI 2025

Multimodal Understanding

Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving

Kewei Zhang^* , Jin Wang^* , Sensen Gao , Chengyue Wu , Yulong Cao , Songyang Han , Boris Ivanovic , Langechuan Liu , Marco Pavone , Song Han , Daquan Zhou , Enze Xie * Equal contribution

Preprint

Fast-dVLM: Efficient Block-Diffusion VLM via Direct Conversion from Autoregressive VLM

Chengyue Wu^* , Shiyi Lan^* , Yonggan Fu , Sensen Gao , Jin Wang , Jincheng Yu , Jose M. Alvarez , Pavlo Molchanov , Ping Luo , Song Han , Ligeng Zhu , Enze Xie * Equal contribution

Preprint

Cross-modal Identity Mapping: Minimizing Information Loss in Modality Conversion via Reinforcement Learning

Haonan Jia^* , Shichao Dong^* , Xin Dong , Zenghui Sun , Jin Wang , Jinsong Lan , Xiaoyong Zhu , Bo Zheng , Kaifu Zhang * Equal contribution

In CVPR 2026

INTER: Mitigating Hallucination in Large Vision-Language Models by Interaction Guidance Sampling

Xin Dong^* , Shichao Dong^* , Jin Wang^* , Jing Huang , Li Zhou , Zenghui Sun , Lihua Jing , Jingsong Lan , Xiaoyong Zhu , Bo Zheng * Equal contribution

In ICCV 2025

Diagnosing the Compositional Knowledge of Vision Language Models from a Game-Theoretic View

Jin Wang , Shichao Dong , Yapeng Zhu , Kelu Yao , Weidong Zhao , Chao Li , Ping Luo

In ICML 2024

Multimodal Evaluation

Forensics-Bench: A Comprehensive Forgery Detection Benchmark Suite for Large Vision Language Models

Jin Wang^* , Chenghui Lv^* , Xian Li , Shichao Dong , Huadong Li , Kelu Yao , Chao Li , Wenqi Shao , Ping Luo * Equal contribution

In CVPR 2025

MMIU: Multimodal Multi-image Understanding for Evaluating Large Vision-Language Models

Fanqing Meng^* , Jin Wang^* , Chuanhao Li^* , Quanfeng Lu , Hao Tian , Jiaqi Liao , Xizhou Zhu , Jifeng Dai , Yu Qiao , Ping Luo , Kaipeng Zhang , Wenqi Shao * Equal contribution

In ICLR 2025

MMT-Bench: A Comprehensive Multimodal Benchmark for Evaluating Large Vision-Language Models Towards Multitask AGI

Kaining Ying^* , Fanqing Meng^* , Jin Wang^* , Zhiqian Li , Han Lin , Yue Yang , Hao Zhang , Wenbo Zhang , Yuqi Lin , Shuo Liu , Jiayi Lei , Quanfeng Lu , Runjian Chen , Peng Xu , Renrui Zhang , Haozhe Zhang , Peng Gao , Yali Wang , Yu Qiao , Ping Luo , Kaipeng Zhang , Wenqi Shao * Equal contribution

In ICML 2024

Deepfake Detection

Towards Understanding the Generalization of Deepfake Detectors from a Game-Theoretical View

Kelu Yao , Jin Wang , Boyu Diao , Chao Li

In ICCV 2023

Implicit Identity Leakage: The Stumbling Block to Improving Deepfake Detection Generalization

Shichao Dong^* , Jin Wang^* , Renhe Ji , Jiajun Liang , Haoqiang Fan , Zheng Ge * Equal contribution

In CVPR 2023

Explaining Deepfake Detection by Analysing Image Matching

Shichao Dong^* , Jin Wang^* , Jiajun Liang , Haoqiang Fan , Renhe Ji * Equal contribution

In ECCV 2022

AI Interpretability

Interpretable Generative Adversarial Networks

Chao Li^* , Kelu Yao^* , Jin Wang^* , Boyu Diao , Yongjun Xu , Quanshi Zhang * Equal contribution

In AAAI 2022 (ORAL)

Jin Wang

CS PhD Student at HKU

The University of Hong Kong

Education

Publications

Unified Multimodal Understanding & Generation

FUDOKI: Discrete Flow-based Unified Understanding and Generation via Kinetic-Optimal Velocities

Multimodal Generation

TAGRPO: Boosting GRPO on Image-to-Video Generation with Direct Trajectory Alignment

Rotate Your Character: Revisiting Video Diffusion Models for High-Quality 3D Character Generation

Megactor-sigma: Unlocking flexible mixed-modal control in portrait animation with diffusion transformer

Multimodal Understanding

Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving

Fast-dVLM: Efficient Block-Diffusion VLM via Direct Conversion from Autoregressive VLM

Cross-modal Identity Mapping: Minimizing Information Loss in Modality Conversion via Reinforcement Learning

INTER: Mitigating Hallucination in Large Vision-Language Models by Interaction Guidance Sampling

Diagnosing the Compositional Knowledge of Vision Language Models from a Game-Theoretic View

Multimodal Evaluation

Forensics-Bench: A Comprehensive Forgery Detection Benchmark Suite for Large Vision Language Models

MMIU: Multimodal Multi-image Understanding for Evaluating Large Vision-Language Models

MMT-Bench: A Comprehensive Multimodal Benchmark for Evaluating Large Vision-Language Models Towards Multitask AGI

Deepfake Detection

Towards Understanding the Generalization of Deepfake Detectors from a Game-Theoretical View

Implicit Identity Leakage: The Stumbling Block to Improving Deepfake Detection Generalization

Explaining Deepfake Detection by Analysing Image Matching

AI Interpretability

Interpretable Generative Adversarial Networks

Additional Research

Asymmetric Cross-Modal Knowledge Distillation: Bridging Modalities with Weak Semantic Consistency

SWIRL: A Staged Workflow for Interleaved Reinforcement Learning in Mobile GUI Control

Falcon: A remote sensing vision-language foundation model

Sparse Beats Dense: Rethinking Supervision in Radar-Camera Depth Completion

Towards RGB-NIR Cross-modality Image Registration and Beyond