Lightnews — Scholar-powered news

Mathilde Papillon @mathildepapillon.bsky.social · 16d

We can't wait to see your innovative solutions! Full details on the tasks and data are on our site:

🔗 geometric-intelligence.github.io/topobench/td...

#TDLChallenge #AI #ML #TAGinDS

3 6

Mathilde Papillon @mathildepapillon.bsky.social · 16d

A huge thank you to our partners! 🤝

💰 Cash prizes are sponsored by the brilliant team at Arlequin AI.

🏛️ Internships are hosted by the world-class UC Santa Barbara @geometric-intel.bsky.social and EPFL IMOS labs.

1 3 5

Mathilde Papillon @mathildepapillon.bsky.social · 16d

🏆The 2025 Topological Deep Learning Challenge is officially live!🏆
Hosted by the TAGinDS Conference 2025

🚀 This year, push the boundaries of topological ML by “Expanding the Data Landscape”

💰 Win incredible prizes, including research internships at UCSB and EPFL, as well as up to $800 US in cash!

1 5 14

Reposted by Mathilde Papillon

Nina Miolane @ninamiolane.bsky.social · Aug 6

In the 20th century, non-Euclidean geometry transformed how we model the world with pen and paper.

Now, it is disrupting how we model the world with machines.

Learn more in our illustrated guide to non-Euclidean ML and AI👇

Mathilde Papillon @mathildepapillon.bsky.social · Aug 1

Our illustrated guide to non-Euclidian ML is finally published!
Check it out for
⭐️ gorgeous figures (with new additions!) on topology, algebra, and geometry in the field
⭐️ broken down tables for easy reading
⭐️ accessible text, additional refs, and more
iopscience.iop.org/article/10.1...

1 6 11

Mathilde Papillon @mathildepapillon.bsky.social · Aug 1

The updated version is also available on arxiv at arxiv.org/pdf/2407.09468

arxiv.org

1 3 9

Mathilde Papillon @mathildepapillon.bsky.social · Aug 1

Thank you to the excellent team of editors @iopp-mlresearch.bsky.social and our referees for a truly productive peer review process. Highly recommend this up and coming journal for any unsubmitted topical reviews out there!

1 2 6

Mathilde Papillon @mathildepapillon.bsky.social · Aug 1

Made possible with the (equal) dedication of @naturecomputes.bsky.social Johan Mathe @louisacornelis.bsky.social and the contributions of @abbybertics.bsky.social , Domas Buracas, Hansen Lillemark, @christian-shewmake.bsky.social , @fatihdinc.bsky.social Xavier Pennec and @ninamiolane.bsky.social

1 2 7

Mathilde Papillon @mathildepapillon.bsky.social · Aug 1

Our illustrated guide to non-Euclidian ML is finally published!
Check it out for
⭐️ gorgeous figures (with new additions!) on topology, algebra, and geometry in the field
⭐️ broken down tables for easy reading
⭐️ accessible text, additional refs, and more
iopscience.iop.org/article/10.1...

1 19 52

Mathilde Papillon @mathildepapillon.bsky.social · Jul 15

Hi! Who is at ICML? Feel free to reach out:)
Our poster session starts tomorrow at 11am E-3101

Mathilde Papillon @mathildepapillon.bsky.social · May 19

TopoTune is going to ICML 2025!🎉🇨🇦
Curious to try topological deep learning with your custom GNN or your specific dataset? We built this for you! Find out how to get started at geometric-intelligence.github.io/topotune/

4

Reposted by Mathilde Papillon

PLOS Biology @plosbiology.org · Jun 11

Can #AI become a true scientist? @ninamiolane.bsky.social explores how new technologies are reshaping scientific discovery, and why human expertise remains essential as we enter a new era of research powered by intelligent algorithms. 🧪‬
plos.io/4kZi0aB

The fifth era of science: Artificial scientific intelligence

Can AI become a true scientist? This Perspective explores how new technologies are reshaping scientific discovery, and why human expertise remains essential as we enter a new era of research powered b...

plos.io

1 4 15

Reposted by Mathilde Papillon

Nina Miolane @ninamiolane.bsky.social · May 29

Pregnancy &motherhood transform the brain in ways we're only beginning to understand🧠

Uncovering these changes w/ @susanacarmona.bsky.social @magdamartinezga.bsky.social @emilyjacobs.bsky.social @grotzinger.bsky.social Kaya Jordan & @cziscience.bsky.social's generous support!

tinyurl.com/38jb7hm6

3 8 29

Reposted by Mathilde Papillon

Francisco Acosta @franciscoacosta.bsky.social · May 23

We are looking for reviewers to serve on the Program Committee of the Symmetry and Geometry in Neural Representations (NeurReps) workshop at NeurIPS 2025! If you are interested in participating, please fill out this form: forms.gle/wgUXHT7Mq71n...

@neurreps.bsky.social

8 20

Mathilde Papillon @mathildepapillon.bsky.social · May 19

Thank you Guillermo Bernárdez, @clabat9.bsky.social @ninamiolane.bsky.social
for making this work possible!
🏠 @geometric-intel.bsky.social @ucsb.bsky.social

5 7

Mathilde Papillon @mathildepapillon.bsky.social · May 19

🍩TopoTune takes any neural network as input and builds the most general TDL model to date, complete with permutation equivariance and unparalleled expressivity.

⚙️ Thanks to its implementation in TopoBench, defining and training these models only requires a few lines of code.

1 5 7

Mathilde Papillon @mathildepapillon.bsky.social · May 19

TopoTune is going to ICML 2025!🎉🇨🇦
Curious to try topological deep learning with your custom GNN or your specific dataset? We built this for you! Find out how to get started at geometric-intelligence.github.io/topotune/

1 9 22

Mathilde Papillon @mathildepapillon.bsky.social · May 13

Come find me @gsp-workshop.bsky.social at MILA all week! Can’t wait to be back in Montreal connecting with the community🕸️
I’ll be speaking on purely graph-based (!) Topological Deep Learning first thing tomorrow at 9am. Let’s kick off this workshop in style🍩

Graph Signal Processing Workshop 2025 @gsp-workshop.bsky.social · Jan 26

We’re absolutely thrilled to have an exciting lineup of speakers this year covering a wide range of topics, including @pseudomanifold.topology.rocks @dom-beaini.bsky.social @mathildepapillon.bsky.social Dhananjay Bhaskar, Naoki Saito!

4 7

Reposted by Mathilde Papillon

Nina Miolane @ninamiolane.bsky.social · Mar 5

How can neural nets extract *interpretable* features from data—& uncover new science?

👉 Discover our mathematical framework tackling this question w/ identifiability theory, compressed sensing, interpretability & geometry!🌐

By @david-klindt.bsky.social @rpatrik96.bsky.social C. O'Neill H Maurer

8 18

Reposted by Mathilde Papillon

David Klindt @david-klindt.bsky.social · Mar 4

🔵 New paper! We explore sparse coding, superposition, and the Linear Representation Hypothesis (LRH) through identifiability theory, compressed sensing, and interpretability. If you’re curious about lifting neural reps out of superposition, this might interest you! 🤓
arxiv.org/abs/2503.01824

From superposition to sparse codes: interpretable representations in neural networks

Understanding how information is represented in neural networks is a fundamental challenge in both neuroscience and artificial intelligence. Despite their nonlinear architectures, recent evidence sugg...

arxiv.org

1 7 24

Reposted by Mathilde Papillon

Nina Miolane @ninamiolane.bsky.social · Mar 3

How does AI memorize information? How does ChatGPT remember your name? How do RNNs keep track of simple numbers?

Our new paper led by Bariscan Kurtkaya & Fatih Dinc explores the geometry of short-term memory in neural networks—revealing limits in how information is stored!🧵
arxiv.org/abs/2502.17433

1 8 14

Reposted by Mathilde Papillon

Nina Miolane @ninamiolane.bsky.social · Feb 25

🤖 New Talk: Geometric Intelligence in Minds & Machines 🧠 Available on Youtube:
www.youtube.com/watch?v=1RtH...

@louisacornelis.bsky.social & I opened the 2025 @neurreps.bsky.social Global Speaker Series !

Thanks @mathildepapillon.bsky.social Chase VDG @btolooshams.bsky.social for hosting!

1 5 12

Reposted by Mathilde Papillon

Louisa Cornelis @louisacornelis.bsky.social · Feb 22

Thanks for hosting this Mathilde!

Mathilde Papillon @mathildepapillon.bsky.social · Feb 21

Thank you to everyone who showed up in person and online! Congrats to @ninamiolane.bsky.social and @louisacornelis.bsky.social for some stellar talks🤩

1 5 11

Mathilde Papillon @mathildepapillon.bsky.social · Feb 21

Thank you to everyone who showed up in person and online! Congrats to @ninamiolane.bsky.social and @louisacornelis.bsky.social for some stellar talks🤩

1 3 8

Reposted by Mathilde Papillon

Bahareh Tolooshams @btolooshams.bsky.social · Feb 19

Join us remotely or in person to hear from @ninamiolane.bsky.social and @louisacornelis.bsky.social on Geometric Intelligence in Minds and Mahcines.

Thanks, @mathildepapillon.bsky.social, for hosting!

Symmetry and Geometry in Neural Representations @neurreps.bsky.social · Feb 19

📢 NeurReps Speaker Series seminar is back!

@ucsb.bsky.social is hosting us for a talk by @ninamiolane.bsky.social and @louisacornelis.bsky.social on Geometric Intelligence in Minds and Machines! 🧠

🗓️This Friday, February 21st, 11 AM PT
Join in person at UCSB (Elings, 1605) or on Zoom!

3 4

Reposted by Mathilde Papillon

Olga Zaghen @olgatticus.bsky.social · Feb 19

Variational Flow Matching goes Riemannian! 🔮

In this preliminary work, we derive a variational objective for probability flows 🌀 on manifolds with closed-form geodesics, and discuss some interesting results.

Dream team: Floor, Alison & Erik (their @ below) 💥

📜 arxiv.org/abs/2502.12981
🧵1/5

1 12 34

Mathilde Papillon @mathildepapillon.bsky.social · Feb 19

I am excited to host the next edition of the NeurReps Global Seminar Series, from @neurreps.bsky.social!
@ninamiolane.bsky.social and @louisacornelis.bsky.social will be speaking on 🧠Geometric Intelligence in Minds and Machines🤖
📅Friday, Feb 21st, 11am PST
Join us at UCSB (Elings,1605) or on Zoom!

8 23