About

I am a final-year Ph.D. student in the ICL group at the School of Artificial Intelligence, Jilin University, advised by Associate Professor Rui Ma. I am also grateful for mentorship from Yilin Wang (Adobe Research) and Zili Yi (Nanjing University).

I received my B.E. from Dalian Maritime University and my M.S. from the School of Computer Science, Jilin University. I am currently interning at Tencent Youtu through the Qingyun Program; previously I interned at Alibaba Amap, Kunlun Tech, Baidu Maps, and JD Digits.

I founded and co-host PaperABC on Bilibili, where we explain recent papers; the channel has grown to over 12,000 subscribers.

Research

My research focuses on building models that manipulate and synthesize visual content in a controllable, personalized way. Core directions include the following.

  • Image editing
  • Style transfer
  • Unified understanding & generation

News

Mar 2026

I joined Tencent Youtu as a research intern through the Qingyun Program, working on unified understanding & generation and post-training.

Oct 2025

I joined Alibaba Amap as an intern, focusing on physical-aware image generation.

May 2025

Our paper OmniStyle was accepted to CVPR 2025.

Feb 2025

Our paper SigStyle was accepted to AAAI 2025.

Internships

Industry experience in reverse chronological order.

  • Tencent Youtu
    Tencent Youtu (腾讯优图) Mar 2026 — Present

    Qingyun Program (青云计划) · Intern

    • Unified understanding & generation
    • Post-training
  • Alibaba Amap
    Alibaba Amap (高德地图) Oct 2025 — Jan 2026

    Alibaba Group · Intern

    Physical-aware image generation

    • Image generation
    • Physical world modeling
  • Kunlun Tech
    Kunlun Tech (昆仑万维) Mar 2024 — Jun 2024

    AI Story · Research algorithm intern

    Multi-subject person image generation.

    • Deep learning
    • Image generation
  • Baidu Maps
    Baidu Maps (百度地图) Mar 2021 — Jun 2021

    Data mining · Algorithm intern

    Real-time bus ETA estimation and optimization; bus departure interval prediction.

    • Machine learning
    • Time series
    • Big data
  • JD Digits
    JD Digits (京东数科) Nov 2020 — Jan 2021

    Data mining · Algorithm intern

    Grey-user information mining and identification systems.

    • Data mining
    • User profiling
    • Risk identification

Publications

First / corresponding author

  1. Learning to Stylize by Learning to Destylize: A Scalable Paradigm for Supervised Style Transfer

    Ye Wang, Zili Yi, Yibo Zhang, Peng Zheng, Xuping Xie, Jiang Lin, Yijun Li, Yilin Wang, Rui Ma

    Destylization-driven stylization: we learn to stylize by learning to destylize, enabling scalable supervised style transfer.

    Paper

  2. OmniStyle: Filtering High Quality Style Transfer Data at Scale

    Ye Wang, Ruiqi Liu, Jiang Lin, Fei Liu, Zili Yi, Yilin Wang, Rui Ma

    CVPR 2025 (CCF-A)

    Large-scale filtered dataset (OmniStyle-1M) and a DiT-based framework for high-resolution instruction- and image-guided style transfer.

    Paper arXiv Project Code

  3. SigStyle: Signature Style Transfer via Personalized Text-to-Image Models

    Ye Wang, Tongyuan Bai, Xuping Xie, Zili Yi, Yilin Wang, Rui Ma

    AAAI 2025 (CCF-A)

    Captures signature style via a personalized diffusion model and hypernetwork; time-aware attention swapping improves content preservation.

    Paper arXiv Project Code

  4. PairHuman: A High-Fidelity Photographic Dataset for Customized Dual-Person Generation

    Ting Pan*, Ye Wang*, Peiguang Jing, Rui Ma, Zili Yi, Yu Liu

    * Co-first authors, equal contribution （共同一作，平等贡献）.

    IEEE TMM (CCF-A)

    Paper arXiv

  5. DP-Adapter: Dual-Pathway Adapter for Boosting Fidelity and Text Consistency in Customizable Human Image Generation

    Ye Wang, Xuping Xie, Lanjun Wang, Zili Yi, Rui Ma

    Graphical Models (CCF-B)

    Paper arXiv

  6. A Training-Free Framework for High-Fidelity Appearance Transfer via Diffusion Transformers

    Shengrong Gu, Ye Wang, Song Wu, Rui Ma, Qian Wang, Lanjun Wang, Zili Yi

    Tech lead

    ICASSP 2026 (CCF-B)

    Paper link TBA

  7. MXM-CLR: A Unified Framework for Contrastive Learning of Multifold Cross-Modal Representations

    Ye Wang, Bowei Jiang, Changqing Zou, Rui Ma

    IJCV under review (CCF-A)

    Paper

  8. SingleDream: Attribute-Driven T2I Customization from a Single Reference Image

    Ye Wang, Ruiqi Liu, Zili Yi, Tieru Wu, Rui Ma

    CVM 2025 (CCF-C)

    Parameter-efficient fine-tuning with a single reference image for attribute-driven text-to-image customization (style, appearance, shape) via a hypernetwork-enhanced approach.

    Paper

Other authors

  1. FreeControl: Efficient, Training-Free Structural Control via One-Step Attention Extraction

    Jiang Lin, Xinyu Chen, Song Wu, Zhiqiu Zhang, Jizhi Zhang, Ye Wang, Qiang Tang, Qian Wang, Jian Yang, Zili Yi

    NeurIPS 2025 (CCF-A)

    Paper arXiv Project

  2. 3D-SSGAN: Lifting 2D Semantics for 3D-Aware Compositional Portrait Synthesis

    Ruiqi Liu, Peng Zheng, Ye Wang, Rui Ma

    Pacific Graphics 2024 (CCF-B)

    Paper arXiv

  3. P2M2-Net: Part-Aware Prompt-Guided Multimodal Point Cloud Completion

    Linlian Jiang, Pan Chen, Ye Wang, Tieru Wu, Rui Ma

    CAD/Graphics 2023 (CCF-C)

    Paper arXiv Code