Molei Tao

BTW: does anyone know how to fix this:

We knew our paper was cited by more than a handful of papers arXiv'ed months ago, but google scholar still shows 0 citation.

Did we do something that triggered a bug?

15.04.2025 15:24 — 👍 0    🔁 0    💬 0    📌 0

Trivialized Momentum Facilitates Diffusion Generative Modeling on Lie Groups The generative modeling of data on manifolds is an important task, for which diffusion models in flat spaces typically need nontrivial adaptations. This article demonstrates how a technique called `tr...

New blog after only 2 years - big thanks to my excellent student @yuchen4975 for the lead.

Blog is based on one of our ICLR'25 papers. arxiv.org/abs/2405.16381

Happy to talk to anyone at Singapore. Plz just DM!

15.04.2025 15:24 — 👍 0    🔁 0    💬 1    📌 0

Diffusion Generative Model, Non-Euclidean Data, and How the Algebraic/Geometric Structure of Lie Groups Helps TL;DR It is challenging but beneficial to create diffusion generative model on manifolds. By algorithmically introducing an auxiliary momentum variable and mathematically “trivializing” it, this task ...

If you know data live on a manifold, you can hardwire this prior knowledge in diffusion model to make generation more accurate & data efficient.

What if there is also a group structure, like in protein design & quantum problem? Use it to do even better -

itsdynamical.github.io/article/2025...

15.04.2025 15:24 — 👍 4    🔁 1    💬 1    📌 0

Fast Solvers for Discrete Diffusion Models: Theory and Applications of High-Order Algorithms Discrete diffusion models have emerged as a powerful generative modeling framework for discrete data with successful applications spanning from text generation to image synthesis. However, their deplo...

How to accelerate the inference of discrete diffusion model / improve its accuracy?
Please check out arxiv.org/abs/2502.00234 .
Grateful to have worked with a luxurious team of
Yinuo Ren, Haoxuan Chen, Yuchen Zhu, Wei Guo, Yongxin Chen, @grant.rotskoff.cc and @lexingying.bsky.social !

07.02.2025 14:27 — 👍 7    🔁 2    💬 0    📌 0

Decision tree that has been sent to Program Officers at NSF

Decision tree that has been sent to Program Officers at NSF

04.02.2025 01:33 — 👍 2778    🔁 899    💬 112    📌 183

Looks to be a really nice read; clear that the man enjoys writing!

arxiv.org/abs/2501.00925
'Fisher Information in Kinetic Theory'
- Cédric Villani

03.01.2025 09:55 — 👍 54    🔁 9    💬 2    📌 2

FutureHouse is launching an independent postdoctoral fellowship program for exceptional researchers who want to apply our automated science tools to specific problems in biology and biochemistry, in collaboration with world-leading academic labs. 1/

19.12.2024 15:29 — 👍 49    🔁 25    💬 1    📌 9

BreimanLectureNeurIPS2024_Doucet.pdf

The slides of my NeurIPS lecture "From Diffusion Models to Schrödinger Bridges - Generative Modeling meets Optimal Transport" can be found here
drive.google.com/file/d/1eLa3...

15.12.2024 18:40 — 👍 327    🔁 67    💬 9    📌 6

I'll be at @neuripsconf.bsky.social presenting 4 papers.

And... anyone wants to chat about:
* diffusion model
* AI4Science (quantum, astronomy, molecule, computing...)
* deep learning theory
* dynamical systems, geometry, stochastic analysis, control, computational math?
Simply DM, reply or email!

06.12.2024 14:25 — 👍 11    🔁 1    💬 0    📌 0

I have updated my course notes on Optimal Transport with a new Chapter 9 on Wasserstein flows. It includes 3 illustrative applications: training a 2-layer MLP, deep transformers, and flow-matching generative models. You can access it here: mathematical-tours.github.io/book-sources...

04.12.2024 08:11 — 👍 103    🔁 19    💬 2    📌 0

NeurIPS Poster Accelerating Diffusion Models with Parallel Sampling: Inference at Sub-Linear Time ComplexityNeurIPS 2024

If you're at NeurIPS next week come see our spotlight poster led by Yinuo Ren and Haoxuan Chen! We use the parallel sampling technique to rigorously establish a big acceleration for diffusion model inference! neurips.cc/virtual/2024...

03.12.2024 21:55 — 👍 9    🔁 2    💬 0    📌 0

Admissions Georgia Tech welcomes you! With over 1,000 universities in the United States that offer graduate degrees, we are honored that you are considering advancing your education at Tech.

GT dynamical ML group is recruiting PhD students for Fall'25, in ML theory and/or AI4Science.

Interested and have Math/Phys/CS background? Please apply at grad.gatech.edu/admissions to either Math, ML, CSE, or ACO PhD program, mention my name & choose Math as home unit.

Repost is appreciated!

27.11.2024 14:58 — 👍 2    🔁 0    💬 0    📌 0