CS PhD student at @MIT • Prev: Stats/Neuro @OxfordUni• Math at @UniBonn • Former: @NVIDIA, @AIatMeta
Joined July 2022
- Tweets102
- Following682
- Followers5K
- Likes206
Pinned Tweet
🚀MIT Flow Matching and Diffusion Lecture 2026 Released (diffusion.csail.mit.edu/)!
We just released our new MIT 2026 course on flow matching and diffusion models! We teach the full stack of modern AI image, video, protein generators - theory and practice. We include:
📺 Videos: Step-by-step derivations.
📝 Notes: Mathematically self-contained lecture notes
💻 Coding: Hands-on exercises for every component
We fully improved last years’ iteration and added new topics: latent spaces, diffusion transformers, building language models with discrete diffusion models.
Everything is available here: diffusion.csail.mit.edu/
A huge thanks to Tommi Jaakkola for his support in making this class possible and Ashay Athalye (MIT SOUL) for the incredible production! Was fun to do this with @RShprints!
#MachineLearning #GenerativeAI #MIT #DiffusionModels #AI
Peter Holderrieth retweeted
I found the first lecture of this (youtu.be/9eJQQVrUUoI) to be the most gentle introduction to flow matching and diffusion models I've seen. Really makes it click.
🚀MIT Flow Matching and Diffusion Lecture 2026 Released (diffusion.csail.mit.edu/)!
We just released our new MIT 2026 course on flow matching and diffusion models! We teach the full stack of modern AI image, video, protein generators - theory and practice. We include:
📺 Videos: Step-by-step derivations.
📝 Notes: Mathematically self-contained lecture notes
💻 Coding: Hands-on exercises for every component
We fully improved last years’ iteration and added new topics: latent spaces, diffusion transformers, building language models with discrete diffusion models.
Everything is available here: diffusion.csail.mit.edu/
A huge thanks to Tommi Jaakkola for his support in making this class possible and Ashay Athalye (MIT SOUL) for the incredible production! Was fun to do this with @RShprints!
#MachineLearning #GenerativeAI #MIT #DiffusionModels #AI
Peter Holderrieth retweeted
How can we improve robot generalization beyond collecting more demonstrations?
A robot can use inference to work out how to solve tasks it was never explicitly trained to perform. With learned models of the world, we can plan our future actions and goals before acting.
I wrote a research perspective, Generalization by Construction, on how learning can be combined with inference for flexible generalization.
yilundu.github.io/perspectiv…
Peter Holderrieth retweeted
"An Introduction to Flow Matching and Diffusion Models" is a set of MIT lecture notes from the 2026 course "Generative AI" that I found useful for understanding the mathematics behind modern generative AI.
The notes build the subject gradually, starting from probability distributions, ordinary and stochastic differential equations, and Brownian motion, before moving to diffusion processes, flow matching, score matching, and classifier-free guidance. They also explore architectures for image and video generation, latent spaces, autoencoders, and discrete diffusion models for language generation.
It's a really interesting resource.
diffusion.csail.mit.edu/2026…
Very elegant distillation method for flow models! This works just out-of-the-box and is both simple and scalable!
1/ New paper: Parallel Decoding Distillation (PDD)!🚀
To accelerate diffusion/flow models, trajectory distillation usually merges consecutive ODE steps into 1 larger step. PDD changes this by predicting MULTIPLE ODE steps in 1 network eval.🧵
w/ @julberner @ArashVahdat Chao Liu
Peter Holderrieth retweeted
1/ New paper: Parallel Decoding Distillation (PDD)!🚀
To accelerate diffusion/flow models, trajectory distillation usually merges consecutive ODE steps into 1 larger step. PDD changes this by predicting MULTIPLE ODE steps in 1 network eval.🧵
w/ @julberner @ArashVahdat Chao Liu
Peter Holderrieth retweeted
In January, I started "building something new" with an incredible team. Today I finally get to share some first details about what we've been building. We've called it Walden Robotics (waldenrobotics.com).
I thought long and hard about my own reasons for starting this company. It's not only about the robots. It's also about people. I've tried to capture those thoughts in my first Walden blog post: waldenrobotics.com/news/why-…
It's been an incredible ride so far. Within just a few months of forming the company, we were already operating a general-purpose robot with an end-to-end policy in production in one of the most important factories in North America. It's amazing at how much I've already learned from that experience. There is a lot of work to do, but the mission has never been so clear.
Please help me welcome Walden Robotics into the world. And stay tuned for more updates!
youtube.com/watch?v=fewvZrck…
Peter Holderrieth retweeted
I‘m at ICML this week and we’re presenting Diamond Maps at Hall A #2713 today at 2pm @electronickale, @zeynepakata, @peholderrieth!
Also, feel free to DM if you wanna chat about diffusion/flows/reward alignment/etc.
Hi everyone! I’ve moved to the Bay Area for a summer research internship at @nvidia. Beyond exciting work, I'd love to meet new people doing exciting stuff (incl. stuff I don't work on myself rn!). If you’re around, I’d love to connect! Even if just for a jam session!
Peter Holderrieth retweeted
🤖 We introduce Ambient Diffusion Policy, a simple and principled method for training policies with suboptimal data in robotics.
Suboptimal data is everywhere in robotics…
❌ Data filtering is wasteful
❌ Co-training learns both good and bad features
✅ Ambient Diffusion Policy selectively learns useful features via noise-dependent data usage
👇🧵(1/5)
Peter Holderrieth retweeted
Big news from Boltz - our biggest update yet! 🚀
Today we’re releasing two new state-of-the-art models for protein and small molecule design with extensive wet lab validation and a new API to run all of our models on scalable GPUs wherever you (or your agents) work! 🔥
Check out our DeCAF model! We and many others promoted for a while that inference-time scaling with flow maps is the way to go. It is super cool to see this working on protein-ligand co-folding models. Excited to see what these ideas will unlock for proteins in the future!
Protein–ligand cofolding models are getting incredibly powerful… but do they have to be so slow? 🧬🐢💊
Our new preprint introduces a new flow-map framework called DeCAF for fast few-step cofolding — up to 5× faster while preserving sample quality on the SOTA Pearl model and 20x faster than Boltz 1x. ⚡🧵
📜 Blog: genesis.ml/news/genesis-mode…
🔗arXiv: arxiv.org/abs/2606.08375
Code (coming soon): github.com/genesistherapeuti…
Peter Holderrieth retweeted
Protein–ligand cofolding models are getting incredibly powerful… but do they have to be so slow? 🧬🐢💊
Our new preprint introduces a new flow-map framework called DeCAF for fast few-step cofolding — up to 5× faster while preserving sample quality on the SOTA Pearl model and 20x faster than Boltz 1x. ⚡🧵
📜 Blog: genesis.ml/news/genesis-mode…
🔗arXiv: arxiv.org/abs/2606.08375
Code (coming soon): github.com/genesistherapeuti…
Wonderful tutorial by @ArashVahdat, @julberner, and @wn8_nie about making diffusion go fast, e.g. enabling cool applications like interactive world models!
From theory to interactive world models ⚡️
The slides for our #CVPR2026 tutorial on accelerated diffusion models are online now!
Topics covered:
1️⃣ General acceleration paradigms (@ArashVahdat)
2️⃣ Step-Distillation (@julberner)
3️⃣ Interactive World Models (@wn8_nie)
Check out the slides on our website cvpr26-tutorial-fastgen.gith… or in the thread below!🧵
Peter Holderrieth retweeted
From theory to interactive world models ⚡️
The slides for our #CVPR2026 tutorial on accelerated diffusion models are online now!
Topics covered:
1️⃣ General acceleration paradigms (@ArashVahdat)
2️⃣ Step-Distillation (@julberner)
3️⃣ Interactive World Models (@wn8_nie)
Check out the slides on our website cvpr26-tutorial-fastgen.gith… or in the thread below!🧵
Peter Holderrieth retweeted
super excited to share our latest work! are we really tilting? 🤨
tldr: reward guidance for flows and diffusions is supposed to sample from the reward-tilted distribution. we show it doesn’t 😰 and how to (mostly) fix it ✨
plus lots of fun images!! 🖼️
collaboration with the awesome @nmboffi
website: sanjitdp.github.io/are-we-re…
paper: arxiv.org/abs/2606.02884
code: github.com/sanjitdp/reward-g…
Peter Holderrieth retweeted
My first blog post in over a year is a deep dive on flow maps🗺️, or how to learn the integral of a diffusion model to enable faster sampling and several other cool tricks.
It's the longest one yet👀 Let me know what you think!
sander.ai/2026/05/06/flow-ma…
Peter Holderrieth retweeted
Up next we have Peter Holderrieth @peholderrieth from MIT talking about glass flows and diamond maps 💎
Very excited about this work! Language modeling via flow maps allows us to rethink how to build language models that do latent reasoning and fast generation!
Glad to play my small part in a wonderful effort with @PPotaptchik, @json_yim, @adhisarav, Eric Vanden-Eijnden, @msalbergo!
New paper! Presenting Discrete Flow Maps:
paper: arxiv.org/abs/2604.09784
blog: malbergo.me/discrete-flow-ma…
A laughable problem for me these days is that @nmboffi and I share a research brain, and we have had, time and again, a conversation that ends with “ha so I guess we’re writing the same paper.” Soon we will return to just doing it together :). Here we are doing it again with discrete flow maps and flow language models! A complete and thorough paper led by @PPotaptchik @json_yim @adhisarav @peholderrieth. We took a bit of time to post it to ensure we understood a few more things about the stability of the loss functions.
Like @osclsd , @FEijkelboom, and @nmboffi , we think this could be a very helpful paradigm for thinking about fast inference and even better alignment!
Here’s our version of the story, and I hope it makes clear how green field this research direction is — we provide a comprehensive picture of the KL losses you can write from the properties of the flow map, some nice geometric proofs about the mean denoiser and the simplex, and find that at this time, the ESD can actually be the most performant, with some caveats. Excited for everyone to work together and push this class of models to their limit!
