Liuyang Bai / Research

Liuyang
Bai

Portrait of Liuyang Bai

I'm a senior undergraduate in Computer Science at NYU Shanghai. My research interest lies in learning compact, semantically structured latent representations of the visual world and such latent representations that allow planning and prediction.

Currently, I'm a research intern at the AI4CE Lab, NYU Tandon, with Prof. Chen Feng. Before that, I was fortunate to work with Prof. Hongyi Wen on flow-based diffusion models, and Prof. Li Guo on low-bitrate speech compression at NYU Shanghai.

I'm looking for CS MS or PhD positions starting Fall 2027.

News

  • This website is live. Hello, world!

  • Our paper on decoder-side semantic conditioning for neural speech compression was accepted to APSIPA ASC 2026.

  • Joined the AI4CE Lab at NYU Tandon, working on latent world models for navigation.

Publications

Scholar ↗
  1. Model overview: a neural speech codec whose decoder is conditioned on quantized semantic features through FiLM layers

    APSIPA ASC 2026

    Decoder-Side Semantic Conditioning for Low-Bitrate Neural Speech Compression

    L. Bai, W. Lu, L. Guo

    [Paper][BibTeX]

  2. CVPR 2026 · Findings

    From Navigation to Refinement: Revealing the Two-Stage Nature of Flow-based Diffusion Models through Oracle Velocity

    H. Liu, J. Liu, Y. Li, L. Bai, Y. Ji, Y. Guo, S. Wan, H. Wen

    [Paper][BibTeX]

  3. EMNLP 2025 · Main

    FinTrust: A Comprehensive Benchmark of Trustworthiness Evaluation in Finance Domain

    T. Hu, T. Hu, L. Bai, Y. Zhao, A. Cohan, C. Zhao

    [Paper][BibTeX]

Photography

All photographs →