Sophia Tang

pic.jpg

sophtang [at] engineering.upenn.edu

Hi! I am a final-year undergraduate student at the University of Pennsylvania conducting research in generative modeling for scientific discovery. Aside from research, I study computer science and statistics in the Jerome Fisher Program in Management & Technology.

Currently, I’m part of the Chatterjee Lab, developing theoretical ML frameworks for biological design, and this summer, I’m a visiting researcher at the Kempner Institute at Harvard University.

My research ranges from developing theoretical Schrödinger bridge frameworks for generative modelling of branching and interacting particle systems to reward alignment and guidance techniques for discrete generative models - but I’m always exploring new theoretical ideas and thinking about interesting problems to apply them to!

I also write long-form tutorial papers on foundational topics in machine learning and shorter technical articles on my Substack, Alchemy Bio. I recently released Foundations of Schrödinger Bridges for Generative Modeling (220 pages) and A Complete Guide to Spherical Equivariant Graph Transformers (99 pages), which aim to break down complex theoretical concepts for a broad audience.

If any of these topics sparks your interest, I would love to connect!


Updates

Jul 24, 2026 Expanding Flow Maps 🌊 is out! We develop a generalizable framework for learning generative flows with dynamically increasing dimensionality!
Jun 15, 2026 A2D2 🃏 and mRNAutilus 🧬 are out! A new framework for reward-alignment of any-length discrete diffusion with adaptive decoding, and an experimentally-validated multi-objective mRNA generative model!
Mar 20, 2026 New tutorial paper on the Foundations of Schrödinger Bridges for Generative Modeling (220 pages, 24 figures) is out on arXiv!
Jan 26, 2026 Branched Schrödinger Bridge Matching is accepted at ICLR 2026 main! 🌳🧫 Looking forward to presenting in Rio! 🇧🇷
Jan 23, 2026 Awarded the CRA Outstanding Undergraduate Researcher Award (1 of 8 across North America)! Read the news article here!
Nov 11, 2025 EntangledSBM is out! Read our paper on simulating interacting multi-particle systems with a novel Schrödinger bridge matching framework!
Sep 30, 2025 TR2-D2 is out! Read our paper on off-policy RL for discrete diffusion fine-tuning with multi-objective rewards!
May 01, 2025 PepTune: De Novo Generation of Therapeutic Peptides with Multi-Objective-Guided Discrete Diffusion accepted at ICML 2025! Come by us at Poster Session 4 to chat (+ free PepTune stickers)

Selected Publications

  1. Sophia Tang and Pranam Chatterjee
    arXiv preprint arXiv:2607.21585, Jul 2026
  2. Sophia Tang , Yuchen Zhu , Molei Tao and Pranam Chatterjee
    arXiv preprint arXiv:2606.13565, Jun 2026
  3. Sophia Tang , Yinuo Zhang and Pranam Chatterjee
    arXiv, Preprint, Nov 2025
  4. Sophia Tang* , Yuchen Zhu* , Molei Tao and Pranam Chatterjee
    arXiv, Preprint, Sep 2025
  5. branchsbm.png
    Sophia Tang , Yinuo Zhang , Alexander Tong and Pranam Chatterjee
    International Conference on Learning Representations (ICLR 2026), Jun 2025
  6. Sophia Tang* , Yinuo Zhang* and Pranam Chatterjee
    International Conference on Machine Learning (ICML 2025), Dec 2024

Invited Talks

  1. July 2026
    Expanding Flows for Fast and Flexible Generation Beyond the Fixed Canvas
    Microsoft Research New England Generative Modeling & Sampling Seminar
  2. July 2026
    Spherical Equivariant Graph Transformers
    Cambridge Ellis Unit Summer School on Probabilistic Machine Learning 2026
  3. Nov 2025
    Peptune: De Novo Generation of Therapeutic Peptides with Guided Discrete Diffusion
    Discrete Diffusion Reading Group
  4. July 2025
    Branched Schrödinger Bridge Matching
    Starkly Speaking Reading Group