Hi, I am Luping Liu (刘路平). I am a third-year Master’s student in Computer Science from Zhejiang University, supervised by Pro. Zhou Zhao. I also collaborate with Yi Ren (Bytedance AI Lab), Chongxuan Li (Renmin University of China) closely. Prior to that, I just obtained my Bachelor’s degree in Mathematics from Chu Kochen Honors College, Zhejiang University, supervised by Pro. Feng Wang. For now, I focus on the acceleration and application of diffusion models, security and generalization of AI. My ultimate goal is to build general artificial intelligence to free people from repetitive and tedious work.

After graduating from Zhejiang University, I will pursue a doctoral degree at the University of Hong Kong with HKUPS! If you are interested in my work, please feel free to contact me via email at luping.liu@zju.edu.cn.

🔥 News

• 2024.02: 🎉🎉 I win the HKU Presidential PhD Scholarship (HKUPS).
• 2023.09: Our PTQD is accepted by NeurIPS 2023.
• 2023.04: 🎉🎉 I get my first 100 citations on Google Scholar.
• 2022.10: 🎉🎉 I win the 2022 national scholarship.
• 2022.05: Our PNDM is used as the default acceleration sampling method for CompVis/stable-diffusion .
• 2022.02: I release the official implementation for PNDM (ICLR 2022).

📝 Publications

ICLR 2022

Pseudo Numerical Methods for Diffusion Models on Manifolds
Luping Liu, Yi Ren, Zhijie Lin, Zhou Zhao

Paper | Project

• This work is a simple, effective and general high-order method for the acceleration of the diffusion model.
• This work is used as the default acceleration sampling method for famous CompVis/stable-diffusion . This work is also included in other famous diffusion model open-source projects, such as huggingface/diffusers and MoonInTheRiver/DiffSinger . In addition, this work is reported by PaperWeekly.

Preprint

Diffusion Denoising Process for Perceptron Bias in Out-of-distribution Detection
Luping Liu, Yi Ren, Xize Cheng, Rongjie Huang, Chongxuan Li, Zhou Zhao

Paper | Project

• This work provides a new perceptron bias assumption for the overconfidence problem and designs two-step diffusion-detection methods for OOD detection.

Preprint

Detector Guidance for Multi-Object Text-to-Image Generation
Luping Liu, Zijian Zhang, Yi Ren, Rongjie Huang, Xiang Yin, Zhou Zhao

Paper | Project

• This work integrates a detection-correction process on cross-attention maps to prevent the amalgamation of multi-object generation.

• ERA-Solver: Error-Robust Adams Solver for Fast Sampling of Diffusion Probabilistic Models. Shengmeng Li, Luping Liu, Zenghao Chai, Runnan Li, Xu Tan. Preprint
• PTQD: Accurate Post-Training Quantization for Diffusion Models. Yefei He, Luping Liu, Jing Liu, Weijia Wu, Hong Zhou, Bohan Zhuang. NeurIPS 2023,
• Make-An-Audio: Text-To-Audio Generation with Prompt-Enhanced Diffusion Models, Rongjie Huang, Jiawei Huang, Dongchao Yang, Yi Ren, Luping liu, Mingze Li, Zhenhui Ye, Jinglin Liu, Xiang Yin, Zhou Zhao. ICML 2023
• Make-A-Voice: Unified Voice Synthesis With Discrete Representation. Rongjie Huang, Chunlei Zhang, Yongqi Wang, Dongchao Yang, Luping Liu, Zhenhui Ye, Ziyue Jiang, Chao Weng, Zhou Zhao, Dong Yu. Preprint

🎖️ Honors and Awards

• 2022.10: National Scholarship (Top 1%).
• 2024.02: HKU Presidential PhD Scholarship.

📖 Educations

• 2021.09 - 2024.03: Master, Zhejiang University, Hangzhou. Supervised by Pro. Zhou Zhao.
• 2017.09 - 2021.06: Undergraduate, Chu Kochen Honors College, Zhejiang Univeristy, Hangzhou. Supervised by Pro. Feng Wang.

💻 Internships

• 2023.07 - 2024.02: Tencent TEG, Shenzhen. Hosted by QingLin Lu and Wei Liu.

If you like the template of this homepage, welcome to star and fork the repo AcadHomepage .