Dar-Yen Chen

Working on visual generation and few-step distillation.

A third-year doctoral candidate at SketchX, Centre for Vision, Speech and Signal Processing, University of Surrey, working with Prof. Yi-Zhe Song.

d.chen@surrey.ac.uk

Publications

  1. Preprint2026

    NitroFusion2: Bringing Adversarial Loss to the Text Stream of Multimodal Diffusion Transformers

    Dar-Yen Chen, Yi-Zhe Song

    arXivProjectCodeDemo

    A frog reading a newspaper headlined NitroFusion2, generated by NitroFusion2
  2. NeurIPS2025

    Normalized Attention Guidance: Universal Negative Guidance for Diffusion Models

    Dar-Yen Chen, Hmrishav Bandyopadhyay, Kai Zou, Yi-Zhe Song

    arXivProjectCode

    Colour sketch of a UFO over a pyramid, generated with Normalized Attention Guidance
  3. CVPR2025

    NitroFusion: High-Fidelity Single-Step Diffusion through Dynamic Adversarial Training

    Dar-Yen Chen, Hmrishav Bandyopadhyay, Kai Zou, Yi-Zhe Song

    arXivProjectCode

    Frost-covered mountain climber portrait, generated by NitroFusion in one step
  4. CVPR2024

    DemoCaricature: Democratising Caricature Generation with a Rough Sketch

    Dar-Yen Chen, Ayan Kumar Bhunia, Subhadeep Koley, Aneeshan Sain, Pinaki Nath Chowdhury, Yi-Zhe Song

    arXivProjectCode

    Caricature generated by DemoCaricature from a rough sketch
  5. CVPR2024

    ArtAdapter: Text-to-Image Style Transfer using Multi-Level Style Encoder and Explicit Adaptation

    Dar-Yen Chen, Hamish Tennent, Ching-Wen Hsu

    arXivProject

    A teacup on a saucer, stylised by ArtAdapter