Pu Cao

Ph.D. student of Artificial Intelligence

Beijing University of Posts and Telecommunications

Biography

Pu Cao is a forth-year Ph.D. student studying at Beijing University of Posts and Telecommunications (BUPT) under the supervision of Prof. Qing Song and Dr. Lu Yang. He is now interested in Computer Vision and am currently working on Visual Synthesis and Multimodal Large-language Model.

Interests

Visual Generation
Multimodal Large Language Models
Embodied AI

Education

PhD in Artificial Intelligence, 2022
Beijing University of Posts and Telecommunications
BSc in Information and Computational Science, 2018
University of Science and Technology Beijing

News

2025.11 One paper is accepted by AAAI 2026 (oral).
2025.01 One paper is accepted by CVPR 2025.

Publications

Browse the highlights below or view the complete publications archive.

Image is All You Need to Empower Large-scale Diffusion Models for In-Domain Generation

CVPR 2025 .

Pu Cao^*, Feng Zhou^*, Lu Yang^✉, Tianrui Huang, Qing Song

Empower diffusion model for in-domain generation.

Image is All You Need to Empower Large-scale Diffusion Models for In-Domain Generation

Exploring Position Encoding in Diffusion U-Net for Training-free High-resolution Image Generation

AAAI 2025 (Oral) .

Feng Zhou^*, Pu Cao^*, Yiyang Ma, Lu Yang, Jianqin Yin^✉

Training-free positional encoding fix for high-resolution diffusion generation.

Exploring Position Encoding in Diffusion U-Net for Training-free High-resolution Image Generation

E4C: Enhance Editability for Text-Based Image Editing by Harnessing Efficient CLIP Guidance

IEEE TCSVT 2025 .

Tianrui Huang^*, Pu Cao^*, Lu Yang, Chun Liu, Mengjie Hu, Zhiwei Liu, Qing Song^✉

Improving editability in text-guided Image editing.

E4C: Enhance Editability for Text-Based Image Editing by Harnessing Efficient CLIP Guidance

Controllable Generation with Text-to-Image Diffusion Models: A Survey

arXiv 2024 .

Pu Cao, Feng Zhou, Qing Song, Lu Yang^✉

A survey on controllable generation with text-to-image diffusion models.

Controllable Generation with Text-to-Image Diffusion Models: A Survey

What Decreases Editing Capability? Domain-Specific Hybrid Refinement for Improved GAN Inversion

WACV 2024 .

Pu Cao, Lu Yang, Dongxv Liu, Xiaoya Yang, Tianrui Huang, Qing Song^✉

Editing capability decreases ineivitably in previous refinement methods, (e.g., PTI, HFGI, and SAM). In this work, we explore the idea of “divide and conquer” to address this problem. We combine two mainstream refinement mechanisms (i.e., weight ande feature modulation) and achieve extroadinary inversion and editing results.