Merging artificial and natural intelligence to study adaptive mechanisms | CV, ML, Neuro | PI: @trackingactions
Joined June 2017
- Tweets 730
- Following 9
- Followers 3,481
- Likes 1,682
73 Photos and videos
Pinned Tweet
π¦π¦π¦ #CEBRA is out πππ
Check out the @Nature paper β¬οΈ & the gorgeous code & resources built by current members @stes_io @BenquetCelia @TrackingActions
@EPFL_en @epflSV
π how to use: cebra.ai/docs/usage.html
π» install: cebra.ai/docs/installation.hβ¦
πnature.com/articles/s41586-0β¦
π¦ Self-supervised multimodal ML is promising the next AI breakthrough - in our new work published in @Nature, we debut @CEBRAai: for self-supervised hypothesis- and discovery-driven science.
π doi.org/10.1038/s41586-023-0β¦
π»github.com/AdaptiveMotorContβ¦
π¦ cebra.ai/
π§΅β¬οΈ
4
16
56
21,580
M- Lab of Adaptive Intelligence @EPFL retweeted
Jun 9
Online now: The Simons Collaboration on Ecological Neuroscience: Studying how the brain interacts with the world dlvr.it/TSyBZX
13
45
3,336
M- Lab of Adaptive Intelligence @EPFL retweeted
π₯³ Excited that our Perspective about our new @SimonsFdn Collaboration: SCENE is out in @NeuroCellPress !
A great team effort across all the PIs, championed by @lengyel_m and @JP__NOEL π
β‘οΈ cell.com/neuron/fulltext/S08β¦
3
16
76
5,942
Interested in how we estimate 3D poses for humans and animals using flow matching?
Come check out our work, #FMPose3D, at CVPR 2026!
Today at the CV4Animals Workshop poster session:
11:30 a.m.β12:30 p.m., ExHall A
Looking forward to seeing you there! π
1
1
3
232
M- Lab of Adaptive Intelligence @EPFL retweeted
π₯³ New work from my lab (@MLabofAI) just dropped on @arxiv! Measuring the shape of an animal can be highly useful for biomedical applications and for avatars π
We introduce *PRIMA*:π₯SOTA 2Dβ‘οΈ3D animal shape by @xiaoyu912huang, @TiwangCS & I!
π arxiv.org/abs/2606.02366
1
13
69
4,488
M- Lab of Adaptive Intelligence @EPFL retweeted
π¨ β¨ New 3D pose estimation method from @mwmathislab! #FMPose3D allows for monocular (i.e. single camera) 2D β‘οΈ 3D π₯
Led by @TiwangCS & w/@xiaoyu912huang #FMPose3D is SOTA on human & animal 3D benchmarks, & will be integrated into @DeepLabCut β¬οΈπ
π arxiv.org/abs/2602.05755
3
16
123
17,079
M- Lab of Adaptive Intelligence @EPFL retweeted
π¨ β¨ New 3D pose estimation method from @mwmathislab! #FMPose3D allows for monocular (i.e. single camera) 2D β‘οΈ 3D π₯
Led by @TiwangCS & w/@xiaoyu912huang #FMPose3D is SOTA on human & animal 3D benchmarks, & will be integrated into @DeepLabCut β¬οΈπ
π arxiv.org/abs/2602.05755
6
42
320
30,493
M- Lab of Adaptive Intelligence @EPFL retweeted
π Finding clinical trials & knowing which are useful is an overwhelming task, even if you know where to start!
I worked on a simple @Gradio app for parsing clinical trials adding a #DeepSeek model to rank trials user Q&A. I hope its useful to many:
clinicaltrialmatcher.org
1
7
32
5,579
M- Lab of Adaptive Intelligence @EPFL retweeted
π» reID of unmarked Grizzly's in Alaskan wilderness from single images is a daunting task, but in our new work out in @CurrentBiology that is exactly what Beth Rosenberg & Mu Zhou & @TrackingPlumes! π₯ Check out the work - many years in the making...
sciencedirect.com/science/arβ¦
1
6
20
1,873
M- Lab of Adaptive Intelligence @EPFL retweeted
π₯πΎ massive congratulations to @shaokaiyeah and Haozhe Qi for #LLaVAction to be accepted to #ICLR2026!
TL;DR the first MMLLM for action understanding, a new benchmark, methods to train your MMLLMs for such tasks πΊπ»ππΌ
π» llavaction.epfl.ch/
π arxiv.org/abs/2503.18712
4
21
2,840
M- Lab of Adaptive Intelligence @EPFL retweeted
π¨ Looking for a primer on the latest advances in joint neural-behavioral modeling?
π€© Check out our new Nature Reviews Neuroscience piece! It was really an honor to write this π«Ά
nature.com/articles/s41583-0β¦
2
23
77
8,721
Nature Neuroscience
Leveraging insights from neuroscience to build adaptive artificial intelligence
nature.com/articles/s41593-0β¦
9
41
3,131
M- Lab of Adaptive Intelligence @EPFL retweeted
π How can we use new neuroscience insights to build *adaptive* AI agents and leverage the many foundation models ( which are much like different brain areas)?
Check out my pitch in @NatureNeuro in the Jan 2026 issue π β¬οΈ
nature.com/articles/s41593-0β¦
#AIagents #neuroscience #AGI
4
25
156
8,776
M- Lab of Adaptive Intelligence @EPFL retweeted
Delighted to have my perspective on how we can leverage the latest insights from neuroscience to build adaptive AI agents out in @NatureNeuro π€π
π nature.com/articles/s41593-0β¦
4
42
4,518
M- Lab of Adaptive Intelligence @EPFL retweeted
30 Dec 2025
Interested in the latest advances in neuroscience (neural dynamics and internal models) and how they can be leveraged to build smarter, adaptive AI?
β‘οΈ My first real solo piece π€π«Ά @NatureNeuro
rdcu.be/eWVmA
5
90
438
27,219
M- Lab of Adaptive Intelligence @EPFL retweeted
10 Dec 2025
The 2025 lab swag / holiday kit! π«π·π§ββοΈ- happy holidays everyone!
2
21
1,488
The Swiss Laboratory Animal Science Association awarded on Wednesday its 2025 Prize to neuroscientist Mackenzie Mathis, Professor at EPFL, βfor her outstanding contribution to the refinement of animal research.βΒ
go.epfl.ch/6689f9
1
4
26
2,732
M- Lab of Adaptive Intelligence @EPFL retweeted
4 Dec 2025
Joint modelling of brain and behaviour dynamics with artificial intelligence β a Review by Mackenzie Weygandt Mathis & Alexander Mathis
@TrackingActions @TrackingPlumes
nature.com/articles/s41583-0β¦
1
19
96
8,637
M- Lab of Adaptive Intelligence @EPFL retweeted
24 Sep 2025
The next OpenSUN3D workshop takes place at @ICCVConference -- we are excited to announce a fantastic line-up of speakers:
π©βπΌ @TrackingActions @sainingxie @jstraub6 @angelxchang @ir0armeni
β±οΈDate/time: Oct. 19, 1:30pm--6pm (1st day, afternoon)
π‘Details: opensun3d.github.io
1
12
73
6,163
The lab hit it out of the park with the 8 year lab anniversary celebration today!!
Extra thanks to @g_roodle for the excellent snacks and science π₯πͺ
1
2
29
2,539
M- Lab of Adaptive Intelligence @EPFL retweeted
25 Aug 2025
My lab has been pushing into explainable, robust, & theoretically-tractable AI models for science πͺ
π¨At #ICCV2025 we introduce #DISTIL - led by amazing PhD student @hsirm96 - we propose a trigger-inversion method for DNNs that reconstructs malicious backdoor triggers 1/4
2
10
38
4,180