Stefan Bauer retweeted
We're releasing dense Argus-format annotations for >5,000 hours (290k episodes) of public robot data (ABC-130k, MolmoAct2 YAM, DROID, GM-100, ABC-Sim), labeled with GPT 6.1 Sol.
Per-hand subgoals and actions, plus outcomes, mistakes, recoveries and object-state changes 👇🧵
6D
Per-hand subgoals and actions, plus outcomes, mistakes, recoveries and object-state changes 👇🧵
6D
1
5
7
153
Stefan Bauer retweeted
NYU Center for Data Science is hiring Faculty Fellows! Apply by Nov 24 apply.interfolio.com/193498
CDS is hiring Faculty Fellows for fall 2027!
Two-year appointments with independent research, research and travel funds, and one course to teach per year.
Apply by November 24 for full consideration: apply.interfolio.com/193498
0
3
8
514
Stefan Bauer retweeted
This project started with a literature search on "growing" work in modern vision pipelines, and I was surprised that there was not much, although there have been more recent works in language. Since I started my PhD by working on vision (object-centric?!) I was interested.
Excited to share “Growing to See”: growing SSL vision models provide better representations with up to 25% fewer FLOPs!
Co-led with @KaplFer.
With @Google’s Paradigms of Intelligence team (@kaitlinmaile & @oswaldjoh), @andrea_dittadi, Karl H. Johansson & @stefanAbauer.
1/13
1
4
9
334
Stefan Bauer retweeted
Excited to share “Growing to See”: growing SSL vision models provide better representations with up to 25% fewer FLOPs!
Co-led with @KaplFer.
With @Google’s Paradigms of Intelligence team (@kaitlinmaile & @oswaldjoh), @andrea_dittadi, Karl H. Johansson & @stefanAbauer.
1/13
Co-led with @KaplFer.
With @Google’s Paradigms of Intelligence team (@kaitlinmaile & @oswaldjoh), @andrea_dittadi, Karl H. Johansson & @stefanAbauer.
1/13
1
11
34
1,508
Stefan Bauer retweeted
🧵 1/10
One-step generation is advancing fast: from flow maps to drifting models.
But how do we train a one-step sampler without target samples? 🤔
Introducing OSIS: “Amortised Inference through One-Step Implicit Sampling”.
Work co-led with @ktamogashev, as part of an ran Carter, @shchoi55, @stefanAbauer and Esmeralda S. Whitammer! 🙌
📄 arxiv.org/abs/2610.12679
More details about the method 🧵👇
One-step generation is advancing fast: from flow maps to drifting models.
But how do we train a one-step sampler without target samples? 🤔
Introducing OSIS: “Amortised Inference through One-Step Implicit Sampling”.
Work co-led with @ktamogashev, as part of an ran Carter, @shchoi55, @stefanAbauer and Esmeralda S. Whitammer! 🙌
📄 arxiv.org/abs/2610.12679
More details about the method 🧵👇
1
7
31
14,294
Stefan Bauer retweeted
🆕 Very happy to share:
“One for All, All for One: Coordinated Multi-Agent Diffusion Steering via Stochastic Optimal Control”
Work co-led with top-notch collaborators @BarbanoRiccardo and Runchang Li 🤝
We show how to teach pretrained diffusion models to work together: , human motion, navigation and image components.
With frozen pretrained weights! ❄️
📄 Paper: arxiv.org/abs/2610.08595
Grateful to an incredible team: George Webber, Alexander Denker, Željko Kereta, @stefanAbauer, Francisco Vargas and Esmeralda S. Whitammer. 🙌
Idea, theory, and some examples below 🧵👇
“One for All, All for One: Coordinated Multi-Agent Diffusion Steering via Stochastic Optimal Control”
Work co-led with top-notch collaborators @BarbanoRiccardo and Runchang Li 🤝
We show how to teach pretrained diffusion models to work together: , human motion, navigation and image components.
With frozen pretrained weights! ❄️
📄 Paper: arxiv.org/abs/2610.08595
Grateful to an incredible team: George Webber, Alexander Denker, Željko Kereta, @stefanAbauer, Francisco Vargas and Esmeralda S. Whitammer. 🙌
Idea, theory, and some examples below 🧵👇
4
9
37
19,541
Stefan Bauer retweeted
@fvderop Incredible spin. Takes real talent to take a rigorous & reasonable study (that should have been done before they built their big foundation models) & twist it to further their virtual cell nonsense propaganda.
2
4
61
3,165
Stefan Bauer retweeted
Best possible model quality under fixed training compute, memory, and enable test time scaling under varying recurrence ?
We systematically study how model size, data, sparsity, and recurrent depth interact.
Loop Scaling Laws predict the frontier: how much effective capacity aturates, and how MoE sparsity raises that ceiling.
Work by @yanbei_c (follow her!)
arxiv.org/abs/2609.40316
We systematically study how model size, data, sparsity, and recurrent depth interact.
Loop Scaling Laws predict the frontier: how much effective capacity aturates, and how MoE sparsity raises that ceiling.
Work by @yanbei_c (follow her!)
arxiv.org/abs/2609.40316
Frontier models like GPT-6 are reportedly using looping. But here're the million-dollar questions on scaling: 🤔 Can adding more loops deliver gains indefinitely? And what happens when looping meets Mixture-of-Experts?
Check out our latest Loop Scaling Laws answering these! 👇
3
10
104
12,621
Stefan Bauer retweeted
📣 Applications are open for the CIFAR Global Scholars program!
Develop and lead highly interdisciplinary research, expand your network and gain valuable leadership training.
🔗 Learn more and apply by November 4: cifar.ca/next-generatio…
Develop and lead highly interdisciplinary research, expand your network and gain valuable leadership training.
🔗 Learn more and apply by November 4: cifar.ca/next-generatio…
0
5
6
1,149
Stefan Bauer retweeted
Thanks to the great feedback we received, we’ve made a few updates to the Foundations of Diffusion manuscript.
Summer isn’t quite over yet, still time to dive into diffusion! 😎🌴
arXiv: arxiv.org/abs/2512.05092
Summer isn’t quite over yet, still time to dive into diffusion! 😎🌴
arXiv: arxiv.org/abs/2512.05092
🚨 Looking for a fully self-contained intro to diffusion models that covers both continuous (images) and discrete (text, sequences) data?
🆕 We just released:
“Foundations of Diffusion Models in General State Spaces: A Self-Contained Introduction”
arxiv : arxiv.org/abs/2512.05092
S/o to @andrea_dittadi for his amazing support & guidance, and huge thanks to @TobiasHppe1, @k_neklyudov, @AlexanderTong7 and @stefanAbauer for their supervision! 🙌
One roadmap for all of diffusion. 🏎️💨
After a few failed posts, broken previews, and getting briefly flagged by X… the full thread's finally out 🤯🧵👇
2
11
62
4,557
Stefan Bauer retweeted
reminder that "physically accurate" video model claims are only as good as the benchmark measuring them! 👩🔬
new Physics-IQ audit found >1/3 of prompts were ambiguous and ~30% of videos had artifacts inflating scores. after cleanup, model rankings changed meaningfully (τ=0.46)
vibes ≠ physics
new Physics-IQ audit found >1/3 of prompts were ambiguous and ~30% of videos had artifacts inflating scores. after cleanup, model rankings changed meaningfully (τ=0.46)
vibes ≠ physics
5
5
19
2,741
Evaluations are the secret sauce of frontier labs and Anates is launching its first project in a collaboration with Deepmind. If you work on video models and or embodied intelligence check out their work! Congrats Tim and Carsten!
𝐂𝐚𝐧 𝐚 𝐯𝐢𝐝𝐞𝐨 𝐦𝐨𝐝𝐞𝐥 𝐚𝐜𝐭𝐮𝐚𝐥𝐥𝐲 𝐫𝐞𝐚𝐬𝐨𝐧 𝐚𝐛𝐨𝐮𝐭 𝐭𝐡𝐞 𝐩𝐡𝐲𝐬𝐢𝐜𝐚𝐥 𝐰𝐨𝐫𝐥𝐝, 𝐨𝐫 𝐢𝐬 𝐢𝐭 𝐣𝐮𝐬𝐭 𝐩𝐫𝐨𝐝𝐮𝐜𝐢𝐧𝐠 𝐩𝐥𝐚𝐮𝐬𝐢𝐛𝐥𝐞-𝐥𝐨𝐨𝐤𝐢𝐧𝐠 𝐦𝐨𝐭𝐢𝐨𝐧?
Today, we are excited to introduce 𝐏𝐡𝐲𝐬𝐢𝐜𝐬-𝐈𝐐 𝐕𝐞𝐫𝐢𝐟𝐢𝐞𝐝
(1/n)
0
1
8
359
Stefan Bauer retweeted
𝐂𝐚𝐧 𝐚 𝐯𝐢𝐝𝐞𝐨 𝐦𝐨𝐝𝐞𝐥 𝐚𝐜𝐭𝐮𝐚𝐥𝐥𝐲 𝐫𝐞𝐚𝐬𝐨𝐧 𝐚𝐛𝐨𝐮𝐭 𝐭𝐡𝐞 𝐩𝐡𝐲𝐬𝐢𝐜𝐚𝐥 𝐰𝐨𝐫𝐥𝐝, 𝐨𝐫 𝐢𝐬 𝐢𝐭 𝐣𝐮𝐬𝐭 𝐩𝐫𝐨𝐝𝐮𝐜𝐢𝐧𝐠 𝐩𝐥𝐚𝐮𝐬𝐢𝐛𝐥𝐞-𝐥𝐨𝐨𝐤𝐢𝐧𝐠 𝐦𝐨𝐭𝐢𝐨𝐧?
Today, we are excited to introduce 𝐏𝐡𝐲𝐬𝐢𝐜𝐬-𝐈𝐐 𝐕𝐞𝐫𝐢𝐟𝐢𝐞𝐝
(1/n)
Today, we are excited to introduce 𝐏𝐡𝐲𝐬𝐢𝐜𝐬-𝐈𝐐 𝐕𝐞𝐫𝐢𝐟𝐢𝐞𝐝
(1/n)
2
9
28
2,259
Stefan Bauer retweeted
AI started in canada with CIFAR, and AI continues here in canada with @CIFAR_News
3
8
126
12,805
Stefan Bauer retweeted
I am moving to @ICComputing at @imperialcollege as an associate professor, where I will be expanding my lab!
I am looking for PhDs and postdocs to join me on my quest to build foundation models with adaptive tokenisation and memory (AToM FMs, funded by @ERC_Research)
I am looking for PhDs and postdocs to join me on my quest to build foundation models with adaptive tokenisation and memory (AToM FMs, funded by @ERC_Research)
21
20
217
16,528
Stefan Bauer retweeted
We are looking forward to welcome @brianltrippe for this months Chalmers AI4Science seminar tomorrow. Check psolsson.github.io/AI4ScienceSemi… for details on how to connect!
0
2
9
1,738
Stefan Bauer retweeted
You really think we're going to scale data labelers to AGI?
Today, we release the largest public long-horizon dataset of human digital work.
600h of long-horizon AGI research across 3 months. 🧵(1/n)
Today, we release the largest public long-horizon dataset of human digital work.
600h of long-horizon AGI research across 3 months. 🧵(1/n)
3
22
166
23,262
Stefan Bauer retweeted
The focus on UMAPs as the output of scRNAseq has debased a whole field.
This fig from arxiv.org/abs/2603.02402 is ridiculous.. showing a faster GPU accelerated workflow.. that produces a completely different UMAP. But since nobody cares what's on the UMAP anyway.. who cares?
This fig from arxiv.org/abs/2603.02402 is ridiculous.. showing a faster GPU accelerated workflow.. that produces a completely different UMAP. But since nobody cares what's on the UMAP anyway.. who cares?
10
40
229
23,185
Stefan Bauer retweeted
ACM has posted an "Expression of Concern" on Igor Markov's article. This stinks to heaven. Apparently, Google did not like the paper. Google is a major donor to ACM. There are also rumors of a legal action by Google against ACM.
cacm.acm.org/research/reeva…
cacm.acm.org/research/reeva…
8
15
112
68,288
Stefan Bauer retweeted
Excited to launch Principia, a nonprofit research organisation at the intersection of deep learning theory and AI safety.
Our goal is to develop theory for modern machine learning that can help us understand network behaviors, including those critical for AI safety.
1
Our goal is to develop theory for modern machine learning that can help us understand network behaviors, including those critical for AI safety.
1
9
35
302
19,372