Qianli Ma
I am a Ph.D. student at Shanghai Jiao Tong University, advised by Prof. Zhipeng Zhang .
I completed my master's degree in Computer Science at Shanghai Jiao Tong University advised by Prof. Li Niu and Prof. Linfeng Zhang and my bachelor's degree in Instrument Science and Control Technology at Southeast University.
I'm interested in generative AI such as diffusion models, LLMs. I am also interested in multimodal learning.
I am actively seeking collaborations on exploring visual generation or reasoning ability of LLMs, please feel free to contact me!!
Google Scholar  / 
GitHub  / 
Twitter  / 
Email
|
|
"What I cannot create, I do not understand."
— Richard Feynman
|
📑 Publications
* denotes equal contribution, † denotes corresponding author, some are highlighted.
|
|
Decouple-Then-Merge: Finetune Diffusion Models as Multi-Task Learning
Qianli Ma, Xuefei Ning, Dongrui Liu, Li Niu†, Linfeng Zhang†
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025
arXiv / PDF / BibTeX / Project Page / Code
This paper proposes a new finetuning method for diffusion models, which decouples the diffusion process into multiple denoising tasks and then merges them. We show that this method can effectively finetune diffusion models for various tasks, including text-to-image generation, unconditional image generation.
|
|
Efficient Diffusion as Low Light Enhancer
Guanzhou Lan*, Qianli Ma*, Yuqi Yang, Zhigang Wang, Dong Wang, Xuelong Li†, Bin Zhao†
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025
arXiv / PDF / BibTeX / Project Page / Code
This paper proposes an efficient diffusion model for low light enhancement, which can be applied to various low light enhancement tasks.
|
|
LED-Merging: Mitigating Safety-Utility Conflicts in Model Merging with Location-Election-Disjoint
Qianli Ma*, Dongrui Liu*, Qian Chen, Linfeng Zhang, Jing Shao†
The 63rd Annual Meeting of the Association for Computational Linguistics (ACL main), 2025
arXiv / BibTeX / Code
This paper proposes a method to mitigate safety-utility conflicts in model merging for LLMs, which can be applied to various safety-utility tasks.
|
|
VBench++: Comprehensive and Versatile Benchmark Suite for Video Generative Models
Ziqi Huang, Fan Zhang, Xiaojie Xu, Yinan He, Jiashuo Yu, Ziyue Dong, Qianli Ma, Nattapol Chanpaisit, Chenyang Si, Yuming Jiang, Yaohui Wang, Xinyuan Chen, Ying-Cong Chen, Limin Wang, Dahua Lin†, Yu Qiao†, Ziwei Liu†
ArXiv, 2024
arXiv / BibTeX / Project Page / Code
This paper proposes a comprehensive and versatile benchmark suite for video generative models.
|
|
Token Pruning for Caching Better: 9 Times Acceleration on Stable Diffusion for Free
Evelyn Zhang, Bang Xiao, Fufu Yu, Jiayi Tang, Chang Zou, Ke Yan, Shouhong Ding, Qianli Ma, Fei Ren, Linfeng Zhang†
ArXiv, 2025
arXiv / BibTeX / Code
This paper proposes a token pruning method for stable diffusion, which can accelerate the generation process by 9 times.
|
|
Shanghai Jiao Tong University
School of Artificial Intelligence
Ph.D. in Artificial Intelligence
Supervised by Prof. Zhipeng Zhang
2025.04 - Now
|
|
Shanghai Jiao Tong University
Department of Computer Science and Engineering
M.Sc. in Computer Science
Supervised by Prof. Li Niu and Prof. Linfeng Zhang
2022.09 - 2025.03
|
|
Southeast University
School of Instrument Science and Engineering
B.Eng. in Instrument Science and Control Technology
2018.09 - 2022.06
|
💻️ Industry and Research Experience
|
|
Baidu
Paddle Team
Machine Learning Engineering Intern
2023.07 - 2023.10
|
🎈 Miscellanea
- Conference Reviewer: ICCV
- Award: First-Class Graduate Academic Scholarship in 2023, 2024
- Award: Freshman Graduate Academic Scholarship in 2022
- Award: Undergraduate Course Scholarship in 2020, 2021
|
|