Pengtao Chen

Ph.D. Student

Embedded Deep Learning and Visual Analysis Lab

College of Future Information Technology, Fudan University (FDU)

Address: 2005 Songhu Rd, Yangpu, Shanghai, China, 200438

E-mail: Pengt.Chen@gmail.com

Citations: GitHub Stars:

Biography []

I am currently a fourth-year Ph.D. student at the College of Future Information Technology, Fudan University, Shanghai, China, supervised by Prof. Tao Chen. I received my B.S. degree from the School of Information Engineering, Zhejiang University of Technology, Hangzhou, China, in 2022, advised by Prof. Qi Xuan. My research interests lie in Diffusion Model, Model Compression, and Efficient Model Design. I am a research intern at the StepFun, working closely with Gang Yu and Xianfang Zeng.

Publications [] [] [] []

(# Equal Contribution; * Corresponding Author.)

2026

CCF-A

ReasonEdit: Towards Reasoning-Enhanced Image Editing Models.

Fukun Yin, Shiyu Liu, Yucheng Han, Zhibo Wang, Peng Xing, Rui Wang, Wei Cheng, Yingming Wang, Aojie Li, Zixin Yin, Pengtao Chen, Xiangyu Zhang, Daxin Jiang^*, Xianfang Zeng^*, Gang Yu^*

Conference on Computer Vision and Pattern Recognition (CVPR), 2026

Preprint

Code

Project

HuggingFace

CCF-A

RegionE: Adaptive Region-Aware Generation for Efficient Image Editing.

Pengtao Chen, Xianfang Zeng, Maosen Zhao, Mingzhu Shen, Peng Ye, Bangyin Xiang, Zhibo Wang, Wei Cheng, Gang Yu, Tao Chen^*

International Conference on Learning Representations (ICLR), 2026

PDF

Preprint

Code

Project

PyPI

CCF-A

Sparse-vDiT: Unleashing the Power of Sparse Attention to Accelerate Video Diffusion Transformers.

Pengtao Chen, Xianfang Zeng, Maosen Zhao, Peng Ye, Mingzhu Shen, Wei Cheng, Gang Yu, Tao Chen^*

AAAI Conference on Artificial Intelligence (AAAI), 2026

Preprint

Code

2025

TechReport

Step1X-Edit: A Practical Framework for General Image Editing.

Led the development of the Diffusers version of Step1X-Edit and optimized its inference efficiency.

Technical Report, 2025

Preprint

Code

Project

Model

Benchmark

CCF-A

DiTFastAttnV2: Head-wise Attention Compression for Multi-Modality Diffusion Transformers.

Hanling Zhang^#, Rundong Su^#, Zhihang Yuan, Pengtao Chen, Mingzhu Shen, Yibo Fan, Shengen Yan, Guohao Dai, Yu Wang^*

International Conference on Computer Vision (ICCV), 2025

PDF

Preprint

Code

CCF-A

Favor-bench: A Comprehensive Benchmark for Fine-Grained Video Motion Understanding.

Chongjun Tu^#, Lin Zhang^#, Pengtao Chen^#, Peng Ye, Xianfang Zeng, Wei Cheng, Gang Yu, Tao Chen^*

Conference on Neural Information Processing Systems (NeurIPS), 2025

PDF

Preprint

Code

Project

HuggingFace

CCF-A

Pioneering 4-Bit FP Quantization for Diffusion Models: Mixup-Sign Quantization and Timestep-Aware Fine-Tuning.

Maosen Zhao^#, Pengtao Chen^#, Chong Yu, Yan Wen, Xudong Tan, Tao Chen^*

Conference on Computer Vision and Pattern Recognition (CVPR), 2025

PDF

Preprint

CCF-A

TSGN: Transaction Subgraph Networks Assisting Phishing Detection in Ethereum.

Jinhuan Wang, Pengtao Chen, Xinyao Xu, Jiajing Wu, Meng Shen, Qi Xuan^*, Xiaoniu Yang

IEEE Transactions on Dependable and Secure Computing (TDSC), 2025

Preprint

IEEE

Code

2024

CCF-A

Δ-DiT: A Training-Free Acceleration Method Tailored for Diffusion Transformers.

Pengtao Chen^#, Mingzhu Shen^#, Peng Ye, Jianjian Cao, Chongjun Tu, Christos-Savvas Bouganis, Yiren Zhao, Tao Chen^*

International Journal of Computer Vision (IJCV), 2024

Preprint

MD-DiT: Step-aware Mixture-of-Depths for Efficient Diffusion Transformers.

Mingzhu Shen^#, Pengtao Chen^#, Peng Ye, Guoxuan Xia, Tao Chen, Christos-Savvas Bouganis, Yiren Zhao^*

NeurIPS Workshop, 2024

PDF

2021

CAS-2

Sampling Subgraph Network with Application to Graph Classification.

Jinhuan Wang, Pengtao Chen, Bin Ma, Jiajun Zhou, Zhongyuan Ruan, Guanrong Chen, Qi Xuan^*

IEEE Transactions on Network Science and Engineering (TNSE), 2021

PDF

Preprint

IEEE

Code

Book

Broad Learning Based on Subgraph Networks for Graph Classification.

Jinhuan Wang, Pengtao Chen, Yunyi Xie, Yalu Shan, Qi Xuan^*, Guanrong Chen

Graph Data Mining Algorithm, Security and Application, 2021

PDF

Springer

CCF-C

TSGN: Transaction Subgraph Networks for Identifying Ethereum Phishing Accounts.

Jinhuan Wang, Pengtao Chen, Shanqing Yu, Qi Xuan^*

International Conference on Blockchain and Trustworthy Systems (BlockSys), 2021

PDF

Preprint

Springer

Internship

StepFun (阶跃星辰), Shanghai, P.R. China

Foundation Model Algorithm Intern (AIGC), from Oct. 2024 to Mar. 2026

Advisor: Gang Yu and Xianfang Zeng.

Focused on instruction-based image editing, video generation and efficient model inference.

Education

College of Future Information Technology, Fudan University, Shanghai, P.R. China

Ph.D., from Sep. 2022 to June 2027 (expected)

Majored in Electronic Science and Technology. Rank: 4/25; GPA: 3.6/4.0.

Jianxing Honors College, Zhejiang University of Technology, Hangzhou, P.R. China

B.Eng., from Sep. 2018 to June 2022

Majored in Automation. Rank: 1/134 (less than 1%); GPA: 4.2/5.0.

Selected Honors & Awards

Nov. 2024 — Fudan University Outstanding Ph.D. Candidate First-Class Scholarship.
Oct. 2023 — Huawei Ascend AI Innovation Competition (ShengSi Algorithm Innovation Track), Gold Award.
Oct. 2023 — The 1st World Science Intelligence Competition (Atmospheric Science Track), Rank: 9/2071.
Feb. 2023 — PCIC 2022 Causal Inference and Recommendation, Rank: 4/161.
May 2022 — Outstanding Graduate of Zhejiang Province.
Dec. 2021 — Zhejiang Provincial Government Scholarship.
Dec. 2020 — National Scholarship (The ONLY student in School of Information Engineering).
Oct. 2020 — The 12th National College Student Mathematics Competition, Second Prize.
Jul 2019 & Sep 2020 — Zhejiang Provincial University Student Mathematics Competition, First Prize × 2.
Oct. 2019 — The 11th National College Student Mathematics Competition, First Prize.
Oct. 2017 — National High School Mathematics League and Chinese Mathematical Olympiad, First Prize (Zhejiang Province).

Academic Service

Conference Reviewer: NeurIPS (2024, 2025), ICLR (2025, 2026), ICML (2025, 2026), CVPR (2025, 2026), ICCV (2025), AAAI (2026), AISTATS (2024).

Journal Reviewer: IJCV, TCSVT.

Collaborators and Friends

Peng Ye (CUHK, Shanghai AI Lab) Mingzhu Shen (IC) Chongjun Tu (FDU) Jianjian Cao (FDU) Lin Zhang (FDU) Jiakang Yuan (FDU) Yaoxin Yang (FDU) Xudong Tan (FDU) Chenyu Huang (FDU) Minglei Li (FDU) Weihao Lin (FDU) Jiong Wang (FDU, Shanghai AI Lab) Hengyuan Xu (FDU) Fukun Yin (StepFun) Wei Cheng (StepFun) Yucheng Han (StepFun) Zhibo Wang (StepFun) Rui Wang (StepFun) Xinyao Liao (NTU) Zhoujie Fu (NTU) Chong Yu (Nvidia) Kun Chen (Shanghai Meteorological Bureau) Xinzhe Zheng (NUS) Hancheng Ye (Duke) Liao Wang (ByteDance) Jiaxu Zhang (ByteDance) Jingyang Xiang (ByteDance) Jinhuan Wang (ZJUT) Junhao Zhu (ZJU) Tian Qiu (ZJU) Bohao Wang (ZJU)