tc.

Apple · Machine Learning Research

Tianrong Chen.

Research Scientist Cupertino, California

I build generative models and study how to make them more capable and more efficient.

My research spans language model pre-training, looped Transformers, and diffusion and flow-based models for language, images, video, and multimodal generation. I am particularly interested in the co-design of algorithms and infrastructure.

Data + infrastructure + co-designed algorithms = better models.

I joined Apple in 2024 after completing my Ph.D. at Georgia Tech, advised by Evangelos A. Theodorou.

Tianrong Chen outdoors beside a bird enclosure
Tianrong ChenResearch Scientist

01 / Research

Learning, generation, efficiency.

01

Language & compute

Large-scale pre-training, looped Transformers, and scaling laws. Parameter-efficient architectures and faster inference for on-device language models.

Language modeling
02

Vision / Multimodal

Normalizing flows, diffusion, and stochastic bridges for image, video, and multimodal generation. Efficient modeling and sampling.

Vision / Multimodal
03

AI for Science

Generative models for scientific discovery, from chemical transition states and quantum states to molecular structure and cellular dynamics.

AI for Science
04

Optimization

Neural optimization, stochastic optimal control, and multi-agent systems. Algorithms informed by dynamic programming and game theory.

Optimization

Recent updates

What’s new

02 / Publications

Research, in print.

Google Scholar

32 publications · * Equal contribution

  1. 2026
    Preprint 2026

    Decoding Looped Transformers Better for (Almost) Free

    Weihao Liu, Huangjie Zheng, Tianrong Chen, Rohit Dilip, Richard He Bai, Yizhu Jiao, Yuyang Wang, Ruixiang Zhang

    Training-free contrastive decoding that halves recurrent loops at matched quality, with 22.5–48.2% fewer forward FLOPs.

    Paper
  2. 2026
    Preprint 2026Mentor

    How to Guide Your Language Flow

    Rohit Dilip, Tianrong Chen, Yuyang Wang, David Van Valen, Josh Susskind, Miguel Angel Bautista

    Guidance for flow-based language generation.

    Paper
  3. 2026
    ICLR 2026

    Continuously Augmented Discrete Diffusion Model for Categorical Generative Modeling

    Huangjie Zheng, Shansan Gong, Ruixiang Zhang, Tianrong Chen, Jiatao Gu, Mingyuan Zhou, Navdeep Jaitly, Yizhe Zhang

    Continuous latent augmentation for discrete diffusion across text, images, and code.

    Paper
  4. 2026
    NeurIPS 2026

    STARFlow2: Bridging Language Models and Normalizing Flows for Unified Multimodal Generation

    Ying Shen, Tianrong Chen, Yuan Gao, Yizhe Zhang, Yuyang Wang, Miguel Angel Bautista, Shuangfei Zhai, Josh Susskind, Jiatao Gu

    Bringing language models and normalizing flows together for unified multimodal generation.

    Paper
  5. 2026
    NeurIPS 2026

    Normalizing Trajectory Models

    Jiatao Gu, Tianrong Chen, Ying Shen, David Berthelot, Shuangfei Zhai, Josh Susskind

    Paper
  6. 2026
    ICML 2026

    Normalizing Flows with Iterative Denoising

    Tianrong Chen, Jiatao Gu, David Berthelot, Josh Susskind, Shuangfei Zhai

    Iterative denoising for end-to-end, likelihood-based normalizing flows.

    Paper Code
  7. 2026
    Preprint 2026

    The Coupling Within: Flow Matching via Distilled Normalizing Flows

    David Berthelot, Tianrong Chen, Jiatao Gu, Marco Cuturi, Laurent Dinh, Bhavik Chandna, Michal Klein, Josh Susskind, Shuangfei Zhai

    Paper
  8. 2026
    CVPR 2026

    One Layer Is Enough: Adapting Pretrained Visual Encoders for Image Generation

    Yuan Gao, Chen Chen, Tianrong Chen, Jiatao Gu

    Paper
  9. 2026
    CVPR 2026

    STARFlow-V: End-to-End Video Generative Modeling with Normalizing Flows

    Jiatao Gu, Ying Shen, Tianrong Chen, Laurent Dinh, Yuyang Wang, Miguel Angel Bautista, David Berthelot, Josh Susskind, Shuangfei Zhai

    Paper Code
  10. 2025
    NeurIPS 2025

    Flexible Language Modeling in Continuous Space with Transformer-based Autoregressive Flows

    Ruixiang Zhang, Shuangfei Zhai, Jiatao Gu, Yizhe Zhang, Huangjie Zheng, Tianrong Chen, Miguel Angel Bautista, Josh Susskind, Navdeep Jaitly

    Language modeling in continuous space with Transformer-based autoregressive flows.

    Paper
  11. 2025
    NeurIPS 2025

    TADA: Improved Diffusion Sampling with Training-free Augmented Dynamics

    Tianrong Chen, Huangjie Zheng, David Berthelot, Jiatao Gu, Josh Susskind, Shuangfei Zhai

    Faster diffusion sampling through training-free augmented dynamics.

    Paper Code
  12. 2025
    NeurIPS 2025

    STARFlow: Scaling Latent Normalizing Flows for High-resolution Image Synthesis

    Jiatao Gu, Tianrong Chen, David Berthelot, Huangjie Zheng, Yuyang Wang, Ruixiang Zhang, Laurent Dinh, Miguel Angel Bautista, Josh Susskind, Shuangfei Zhai

    Scaling normalizing flows to high-resolution image synthesis.

    Paper
  13. 2025
    ICML 2025

    Normalizing Flows are Capable Generative Models

    Shuangfei Zhai, Ruixiang Zhang, Preetum Nakkiran, David Berthelot, Jiatao Gu, Huangjie Zheng, Tianrong Chen, Miguel Angel Bautista, Navdeep Jaitly, Josh Susskind

    Scalable Transformer-based normalizing flows for image generation.

    Paper Code
  14. 2025
    IEEE TPAMI 2025

    Optimal Control Theoretic Neural Optimizer: From Backpropagation to Dynamic Programming

    Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou

    Paper
  15. 2025
    NeurIPS 2025

    Deep Generalized Schrödinger Bridges: From Image Generation to Solving Mean-Field Games

    Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou

    Paper
  16. 2025
    ICLR 2025

    Trivialized Momentum Facilitates Diffusion Generative Modeling on Lie Groups

    Tianrong Chen*, Yuchen Zhu*, Lingkai Kong, Evangelos A. Theodorou, Molei Tao

    Paper
  17. 2025
    Nature Machine Intelligence 2025

    Optimal Transport for Generating Transition States in Chemical Reactions (React-OT)

    Chenru Duan, Guan-Horng Liu, Yuanqi Du, Tianrong Chen, Qiyuan Zhao, Haojun Jia, Carla P. Gomes, Evangelos A. Theodorou, Heather J. Kulik

    Paper
  18. 2024
    ICLR 2024Oral

    Generative Modeling with Phase Stochastic Bridges

    Tianrong Chen, Jiatao Gu, Laurent Dinh, Evangelos A. Theodorou, Josh Susskind, Shuangfei Zhai

    Efficient generative modeling through stochastic dynamics in phase space.

    Paper
  19. 2024
    ICLR 2024

    A Robust Differential Neural ODE Optimizer

    Panagiotis Theodoropoulos, Guan-Horng Liu, Tianrong Chen, Augustinos D. Saravanos, Evangelos A. Theodorou

    Paper
  20. 2024
    Preprint 2024

    Quantum State Generation with Structure-Preserving Diffusion Model

    Yuchen Zhu, Tianrong Chen, Evangelos A. Theodorou, Xie Chen, Molei Tao

    Paper
  21. 2023
    NeurIPS 2023

    Deep Momentum Multi-Marginal Schrödinger Bridge

    Tianrong Chen, Guan-Horng Liu, Molei Tao, Evangelos A. Theodorou

    Paper
  22. 2023
    NeurIPS 2023

    Mirror Diffusion Models for Constrained and Watermarked Generation

    Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou

    Paper
  23. 2023
    Preprint 2023

    Augmented Bridge Matching

    Valentin De Bortoli, Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou, Weilie Nie

    Paper
  24. 2023
    ICML Frontiers4LCD Workshop 2023

    Game Theoretic Neural ODE Optimizer

    Panagiotis Theodoropoulos, Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou

    Paper
  25. 2022
    ICLR 2022

    Likelihood Training of Schrödinger Bridge using Forward-Backward SDEs Theory

    Tianrong Chen*, Guan-Horng Liu*, Evangelos A. Theodorou

    A forward-backward SDE framework for likelihood training of Schrödinger bridges.

    Paper
  26. 2022
    NeurIPS 2022Oral

    Deep Generalized Schrödinger Bridge

    Guan-Horng Liu, Tianrong Chen, Oswin So, Evangelos A. Theodorou

    Paper
  27. 2022
    IEEE CDC 2022

    Deep Graphic FBSDEs for Opinion Dynamics Stochastic Control

    Tianrong Chen, Ziyi Wang, Evangelos A. Theodorou

    Paper
  28. 2021
    NeurIPS 2021Spotlight

    Second-Order Neural ODE Optimizer

    Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou

    Paper
  29. 2021
    ICML 2021Spotlight

    Large-Scale Multi-Agent Deep FBSDEs

    Tianrong Chen, Ziyi Wang, Ioannis Exarchos, Evangelos A. Theodorou

    Paper Code
  30. 2021
    ICML 2021Oral

    Dynamic Game Theoretic Neural Optimizer

    Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou

    Paper
  31. 2021
    ICLR 2021Spotlight

    DDPNOpt: Differential Dynamic Programming Neural Optimizer

    Guan-Horng Liu, Tianrong Chen, Evangelos A. Theodorou

    Paper
  32. 2020
    L4DC 2020

    Feynman-Kac Neural Network Architectures for Stochastic Control Using Second-Order FBSDE Theory

    Marcus A. Pereira*, Ziyi Wang*, Tianrong Chen*, Emily Reed, Evangelos A. Theodorou

    Paper

Let’s connect

Ideas are better shared.

If you’re interested in diffusion models or looped Transformers, let’s connect.

tianrongchen11111@gmail.com