Max Simchowitz

116 posts

Max Simchowitz

Max Simchowitz

@max_simchowitz

Assistant Professor @mldcmu. Formerly: Postdoc @MITEECS, PhD @Berkeley_EECS, Math Undergrad @Princeton. New to Twitter. https://t.co/67bMOAyqK6

Katılım Mayıs 2024
147 Takip Edilen2K Takipçiler
Sabitlenmiş Tweet
Max Simchowitz
Max Simchowitz@max_simchowitz·
⏰⏰(another) science of robot learning paper. Why does action chunking work so well in robotic manipulation? Probably lots of reasons. but here’s one you may not have thought of: control stability. After months of polishing and 5 revisions, check out“ Action Chunking and Exploratory Data Collection Yield Exponential Benefits for Imitation Learning”. The title says it all. Lead by the eloquent @ThomasTCKZhang , illustrated by the talented @dpfroms , supported by @ChaoyiPan of MIP fame (simchowitzlabpublic.github.io/much-ado-about…) fame, and in collaboration with the one and only @NikolaiMatni .
Thomas Zhang@ThomasTCKZhang

🤖🤖Very excited to finally share our new work “Action Chunking and Exploratory Data Collection Yield Exponential Improvements in Behavior Cloning for Continuous Control” Everyone in robotics does action-chunking, but why does it actually work?🤔🤔And, what can theory tell us about the properties of data we should be collecting for robotic behavior cloning? 🧵1/N

English
2
17
168
21.5K
Max Simchowitz
Max Simchowitz@max_simchowitz·
🔔Please submit to the Post-Training for Robotics Foundations Models workshop! And consider participating in our Real-World RL challenge.
Shiduo Zhang@Joey_zh_

#RSS2026 Call for participants 📢 Excited to announce our RSS 2026 Workshop: Post-Training for Robotics Foundation Models, together with the first Real-World Reinforcement Learning Challenge! The workshop is held on July 13 in Sydney. posttraining-for-robotics.github.io。

English
2
3
33
5.7K
Max Simchowitz
Max Simchowitz@max_simchowitz·
If I replace the "anonymous" in my NeurIPS submission with Claude as the sole author, will I get desk rejected?
English
5
0
99
18.7K
Max Simchowitz
Max Simchowitz@max_simchowitz·
Unlocking test-time scaling and search in generative models has been a fundamental obstacle, and is increasingly important as these methods are deployed in science, engineering, and robotics. It was a pleasure to work with @peholderrieth on a paradigm that finally makes this capability genuinely possible, and lets you trade off training time vs inference time according to your needs. Check it out!
Peter Holderrieth@peholderrieth

We release Diamond Maps💎 unlocking accurate and efficient guidance for diffusion models. Our experiments show that our methods scale incredibly well. Excited to see what people will build with this! Accurate guidance has been a notoriously hard problem, but in this work, we’re bringing TWO (!) solutions to the table. The recipe for success: 1️⃣ Speed: Use distilled models (flow maps, mean flows, consistency models). 2️⃣ Exploration: Inject stochasticity to properly explore your search space. Because this fundamentally improves anything using flow matching and diffusion, we see a lot of potential for applications across audio, robotics, molecules, and beyond. Paper: arxiv.org/abs/2602.05993 Code: github.com/PeterHolderrie… Huge thanks to an amazing team: Douglas Chen, @LucaEyring, @ishin_shah, Giri Anantharaman, @electronickale, @zeynepakata, Tommi Jaakkola, @nmboffi, and @max_simchowitz. It was awesome bringing this to life together!

English
0
9
58
13.6K
Max Simchowitz
Max Simchowitz@max_simchowitz·
⚠️Public Service Announcement for ICLR folks headed to Rio: Rio is beautiful, but getting a travel adapter in Brazil is a pain in the ass, and they don't use similar outlets to other major regions (US, UK, Europe, China). Buy a Brazil-specific adapter before traveling. Enjoy :)
English
2
10
89
13.2K
Max Simchowitz retweetledi
Aran Nayebi
Aran Nayebi@aran_nayebi·
1/ As AI agents become increasingly capable, what must *inevitably* emerge inside them? We prove selection theorems: strong task performance forces world models, belief-like memory and—under task mixtures—persistent variables resembling core primitives associated with emotion.
Aran Nayebi tweet media
English
3
37
208
33.5K
Max Simchowitz
Max Simchowitz@max_simchowitz·
Great work by @nmboffi and @AdtRaghunathan and others! Excited to see the power of few-step flow maps extend from images to other domains like text. Check out Nick's post to learn more.
Nicholas Boffi@nmboffi

We just brought flow maps to language modeling for one-step sequence generation 💥 Discrete diffusion is not necessary -- continuous flows over one-hot encodings achieve SoTA performance and ≥8.3× faster generation 🔥 We believe this is a major step forward for discrete generative modeling and language modeling alike. 🚀 Full thread from first author @chandavidlee: x.com/chandavidlee/s…

English
1
4
24
5.8K
Max Simchowitz retweetledi
Aviral Kumar
Aviral Kumar@aviral_kumar2·
Can just a 4B model solve IMO-level proof problems at the level of much stronger LLMs like Gemini 3 Pro? Yes, if you can train the LLM to scale test-time compute well! We're very excited to release our 4B model "QED-Nano", built via an awesome open collab! Details below🧵⬇️
Aviral Kumar tweet media
English
8
26
167
21.9K
Max Simchowitz retweetledi
Boyuan Chen
Boyuan Chen@BoyuanChen0·
Introducing Large Video Planner (LVP-14B) — a robot foundation model that actually generalizes. LVP is built on video gen, not VLA. As my final work at @MIT, LVP has all its eval tasks proposed by third parties as a maximum stress test, but it excels!🤗 boyuan.space/large-video-pl…
English
23
97
577
97.7K
Max Simchowitz retweetledi
Yutong (Kelly) He
Yutong (Kelly) He@electronickale·
I’ve always seen twitter bros be like “keep building keep shipping” when using ai agents for coding, but now that I’ve tried it for real, it feels less like grinding and more adjacent to eating popcorn while bed rotting and watching heated rivalry (I’m not complaining)
English
1
1
38
5.3K
Max Simchowitz retweetledi
Francesco Bertolotti
Francesco Bertolotti@f14bertolotti·
The authors ask whether an N-layer ViT can be rewritten using just K<<N layers by recurring on them. Remarkably, they match DINOv2 performance with only 2-3 layers. The paper also offers rich dynamical-systems analysis. Very cool work! 🔗arxiv.org/abs/2512.19941
Francesco Bertolotti tweet mediaFrancesco Bertolotti tweet mediaFrancesco Bertolotti tweet mediaFrancesco Bertolotti tweet media
English
8
66
654
43K
Max Simchowitz
Max Simchowitz@max_simchowitz·
Standard training of generative policies is surprisingly bad at learning multi-modal strategies from training data. So what can we do to pre-train for diverse solutions and effective post-training? Check out @ajwagenmaker 's new paper for an answer :)
Andrew Wagenmaker@ajwagenmaker

How should we pretrain a policy from demonstrations to ensure it is an effective initialization for RL finetuning, while preserving the performance of the pretrained policy itself? We propose Posterior Behavioral Cloning (PostBC)! (1/11)

English
1
5
64
12K
Max Simchowitz
Max Simchowitz@max_simchowitz·
My friends @elvisnava and @mimicrobotics (makers of awesome robot hands 🧤) just put out a video-first VLA. 📹📹 Motivation: A bitter-lesson alternative to retargeting, and a path to cross-embodiment. Key Idea: Nvidia Cosmos backbone with T5 text encoder, and but pre-train from RGB rather than low-dim actions. Low-Dim actions only needed for a lightweight flow-decoder for task-specific fine-tuning. Benefits: Opens to the door for human data collect with off-the-shelf cloth gloves (and maybe soon, YouTube!), leverages internet scale video, embodiment agnostic. Excited to see where you go next!
Elvis Nava@elvisnavah

Today @mimicrobotics and friends are excited to share mimic-video, a new class of Video-Action Model that elevates video model backbones as first class citizens for robot learning!

English
4
14
114
9K
Max Simchowitz retweetledi
Jonathan Fine
Jonathan Fine@jonathanbfine·
the mistake so many people make is seeing university professors as intellectuals when they’re actually employees at a combination hedge fund and healthcare conglomerate that operates a small luxury resort/sports franchise where student-customers occasionally take classes
English
241
2.8K
25.2K
576.7K
Sander Dieleman
Sander Dieleman@sedielem·
Overgeneralisation is probably the main reason why we're still so reliant on it. I think we are also intentionally collapsing the distribution towards its more canonical examples (guidance is just one way to do that, various forms of post-training are also used for this). The data distribution used in pre-training is usually too broad compared to what we want the model to produce. People don't tend to show direct comparisons anymore these days, but most models are pretty bad without guidance (see x.com/sedielem/statu…). I like the toy experiments in the autoguidance paper: arxiv.org/abs/2406.02507
Sander Dieleman tweet media
English
2
7
107
5K
Max Simchowitz
Max Simchowitz@max_simchowitz·
What is the best current explanation for why CFG is needed for conditional generation? Is it just that are unguided models “undersharpened”/ over smoothed bc of NN inductive bias ?
English
3
5
73
11.7K

Keşfet

@peholderrieth @nmboffi @AdtRaghunathan @MIT @ajwagenmaker @elvisnava @mimicrobotics @sedielem