Lightnews — Scholar-powered news

Reposted by Guillaume Lajoie

Nanda H Krishna @nandahkrishna.bsky.social · 24d

Excited to share that POSSM has been accepted to #NeurIPS2025! See you in San Diego 🏖️

Avery HW Ryoo @averyryoo.bsky.social · Jun 6

New preprint! 🧠🤖

How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?

We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!

🧵1/7

1 3 10

Reposted by Guillaume Lajoie

Blake Richards @tyrellturing.bsky.social · 26d

The CTRL-Labs decoding model paper is out! Saw this presented at Cosyne this year, very cool to see it out.

I would say this is the clearest demonstration of scaling laws in neural decoding to-date.

www.nature.com/articles/s41...

🧠📈 🧪

A generic non-invasive neuromotor interface for human-computer interaction - Nature

A high-bandwidth neuromotor interface offers performant out-of-the-box generalization across people.

www.nature.com

1 6 34

Reposted by Guillaume Lajoie

Henry Farrell @himself.bsky.social · Aug 18

www.programmablemutter.com/p/large-lang... Gopnikism, interactionism, structuralism and role play.

Large language models are cultural technologies. What might that mean?

Four different perspectives

www.programmablemutter.com

5 19 51

Guillaume Lajoie @glajoie.bsky.social · Aug 19

Compositionality is a central desideratum for intelligent systems...but it's a fuzzy concept and difficult to quantify. In this blog post, lab member @ericelmoznino.bsky.social outlines ideas toward formalizing it & surveys recent work. A must-read for interested researchers in AI and Neuro

Eric Elmoznino @ericelmoznino.bsky.social · Aug 18

Very excited to release a new blog post that formalizes what it means for data to be compositional, and shows how compositionality can exist at multiple scales. Early days, but I think there may be significant implications for AI. Check it out! ericelmoznino.github.io/blog/2025/08...

Defining and quantifying compositional structure

What is compositionality? For those of us working in AI or cognitive neuroscience this question can appear easy at first, but becomes increasingly perplexing the more we think about it. We aren’t shor...

ericelmoznino.github.io

5 21

Reposted by Guillaume Lajoie

SilicoLabs @silicolabs.bsky.social · Jul 17

🎉 We’re featured by @mila-quebec.bsky.social for our work on immersive, real-world cognitive science. With LABO, researchers can run full XR experiments—no code needed, real behaviour captured. Special thanks to @tyrellturing.bsky.social for being with us from the start!
tinyurl.com/yc4wpp3t

Modernizing Cognitive Health Research with SilicoLabs | Mila

Cognitive health research is a field dedicated to understanding how our brains work when we think, remember, and learn as we go through life. When combined with medical research and rehabilitation…

mila.quebec

1 2 3

Guillaume Lajoie @glajoie.bsky.social · Jun 6

Excited to share recent progress on foundation-like models for neural data. As many use cases for generalizable models demand flexible online deployment, here we focus on a design enabling low latency real time use. We use hybrid SSM architecture & demonstrate various transfer learning capabilities

Avery HW Ryoo @averyryoo.bsky.social · Jun 6

New preprint! 🧠🤖

How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?

We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!

🧵1/7

12

Reposted by Guillaume Lajoie

Avery HW Ryoo @averyryoo.bsky.social · Jun 6

New preprint! 🧠🤖

How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?

We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!

🧵1/7

2 23 52

Reposted by Guillaume Lajoie

Santa Fe Institute @sfiscience.bsky.social · May 13

Reserve your free tickets to SFI's upcoming Community Lecture! lensic.org/events/blais...

Blaise Agüera y Arcas’ presents 'Computing, Life, and Intelligence' at the Lensic on 🗓️ May 20, 7:30pm MT in-person or online.

3 9 29

Reposted by Guillaume Lajoie

Shahab Bakhtiari @shahabbakht.bsky.social · May 14

Check out our new paper!
Vision models often struggle with learning both transformation-invariant and -equivariant representations at the same time.

@hafezghm.bsky.social shows that self-supervised prediction with proper inductive biases achieves both simultaneously. (1/4)
#MLSky #NeuroAI

Hafez Ghaemi @hafezghm.bsky.social · May 14

Preprint Alert 🚀

Can we simultaneously learn transformation-invariant and transformation-equivariant representations with self-supervised learning?

TL;DR Yes! This is possible via simple predictive learning & architectural inductive biases – without extra loss terms and predictors!

🧵 (1/10)

2 13 45

Reposted by Guillaume Lajoie

Ezekiel Williams @ezekielwilliams.bsky.social · Apr 22

come see @glajoie.bsky.social presenting our poster this week @iclr-conf.bsky.social ! It will be poster #56 in poster session #3. This work was a collaboration between a bunch of us from @mila-quebec.bsky.social and Luca Mazzucato @neuroai-uoregon.bsky.social :)

Ezekiel Williams @ezekielwilliams.bsky.social · Mar 25

1/7: Super excited to share our new paper! This one should be of interest to neuroscientists and deep learning theory folks. This paper was a collaboration with Alexandre Payeur, @averyryoo.bsky.social, Thomas Jiralerspong, @mattperich.bsky.social, Luca Mazzucato, @glajoie.bsky.social

3 11

Guillaume Lajoie @glajoie.bsky.social · Apr 11

❤️‍🩹

1

Reposted by Guillaume Lajoie

Sarthak Mittal @sarthmit.bsky.social · Feb 28

🚨 New Preprint! 🚨

We explore Amortized In-Context Bayesian Posterior Estimation with Niels, @glajoie.bsky.social, Priyank Jaini & @marcusabrubaker.bsky.social ! 🔥

Amortized Conditional Modeling = key to success in large-scale models! We use it to estimate posteriors 🔑

📄 arxiv.org/abs/2502.06601

Amortized In-Context Bayesian Posterior Estimation

Bayesian inference provides a natural way of incorporating prior beliefs and assigning a probability measure to the space of hypotheses. Current solutions rely on iterative routines like Markov Chain ...

arxiv.org

1 1 2

Reposted by Guillaume Lajoie

Sarthak Mittal @sarthmit.bsky.social · Feb 28

🚀 New Preprint! 🚀

In-Context Parametric Inference: Point or Distribution Estimators?

Thrilled to share our work on inferring probabilistic model parameters explicitly conditioned on data, in collab with @yoshuabengio.bsky.social, Nikolay Malkin & @glajoie.bsky.social!

🔗 arxiv.org/abs/2502.11617

In-Context Parametric Inference: Point or Distribution Estimators?

Bayesian and frequentist inference are two fundamental paradigms in statistical estimation. Bayesian methods treat hypotheses as random variables, incorporating priors and updating beliefs via Bayes' ...

arxiv.org

1 2 3

Reposted by Guillaume Lajoie

Erica Busch @elbusch.bsky.social · Apr 3

New preprint! Excited to share our latest work “Accelerated learning of a noninvasive human brain-computer interface via manifold geometry” ft. outstanding former undergraduate Chandra Fincke, @glajoie.bsky.social, @krishnaswamylab.bsky.social, and @wutsaiyale.bsky.social's Nick Turk-Browne 1/8

Accelerated learning of a noninvasive human brain-computer interface via manifold geometry

Brain-computer interfaces (BCIs) promise to restore and enhance a wide range of human capabilities. However, a barrier to the adoption of BCIs is how long it can take users to learn to control them. W...

doi.org

2 20 66

Reposted by Guillaume Lajoie

Avery HW Ryoo @averyryoo.bsky.social · Apr 4

Very late, but had a 🔥 time at my first Cosyne presenting my work with @nandahkrishna.bsky.social, Ximeng Mao, @mattperich.bsky.social, and @glajoie.bsky.social on real-time neural decoding with hybrid SSMs. Keep an eye out for a preprint (hopefully) soon 👀

#Cosyne2025 @cosynemeeting.bsky.social

2 6 31

Guillaume Lajoie @glajoie.bsky.social · Apr 3

Fresh updates on our efforts to understand the effects of online error manipulation during learing. Turns out learning a task with assistive devices (think training wheels) changes how credit assignment mechanisms shapes neural representations in the brain.

Amy Orsborn @neuroamyo.bsky.social · Apr 3

Hot off the presses: big update to our work looking at how adaptive decoders influence neural representations.

We added heroic analyses to show in both experiments & models that the structure of what the brain learns is altered by adaptive decoders. Check it out: www.biorxiv.org/content/10.1...

8

Reposted by Guillaume Lajoie

Ezekiel Williams @ezekielwilliams.bsky.social · Mar 25

1/7: Super excited to share our new paper! This one should be of interest to neuroscientists and deep learning theory folks. This paper was a collaboration with Alexandre Payeur, @averyryoo.bsky.social, Thomas Jiralerspong, @mattperich.bsky.social, Luca Mazzucato, @glajoie.bsky.social

6 14 69

Guillaume Lajoie @glajoie.bsky.social · Mar 25

If you'll be at COSYNE workshops, we got a capstone party planned !

Blake Richards @tyrellturing.bsky.social · Mar 24

Coming to the #Cosyne2025 workshops? Wanna dance on the final night? We got you covered.

@glajoie.bsky.social and I have organized a party in Tremblant. Come and get on the dance floor y'all. 🕺

April 1st
10PM-3AM
Location: Le P'tit Caribou

DJs Mat Moebius, Xanarelle, and Prosocial

Please share!

Party poster for dance party on final night of Cosyne 2025 workshops. It will take place April 1st, 2025, 10PM to 3AM at Le P'tit Caribou.

4

Reposted by Guillaume Lajoie

Blake Richards @tyrellturing.bsky.social · Mar 24

Coming to the #Cosyne2025 workshops? Wanna dance on the final night? We got you covered.

@glajoie.bsky.social and I have organized a party in Tremblant. Come and get on the dance floor y'all. 🕺

April 1st
10PM-3AM
Location: Le P'tit Caribou

DJs Mat Moebius, Xanarelle, and Prosocial

Please share!

2 10 43

Guillaume Lajoie @glajoie.bsky.social · Feb 28

As sequence models and in-context conditioning for inference are being developed to perform all kinds of ML tasks, we make systematic and tracktable evaluations to compare point v.s. distributional estimates . imo a key step to scale predictive modeling for general ML

Sarthak Mittal @sarthmit.bsky.social · Feb 28

🚀 New Preprint! 🚀

In-Context Parametric Inference: Point or Distribution Estimators?

Thrilled to share our work on inferring probabilistic model parameters explicitly conditioned on data, in collab with @yoshuabengio.bsky.social, Nikolay Malkin & @glajoie.bsky.social!

🔗 arxiv.org/abs/2502.11617

In-Context Parametric Inference: Point or Distribution Estimators?

Bayesian and frequentist inference are two fundamental paradigms in statistical estimation. Bayesian methods treat hypotheses as random variables, incorporating priors and updating beliefs via Bayes' ...

arxiv.org

3

Reposted by Guillaume Lajoie

Mila - Institut québécois d'IA @mila-quebec.bsky.social · Feb 20

This week, we’re unveiling two members for the AI Insights for Policymakers program: @glajoie.bsky.social (Mila) and Laleh Seyyed-Kalantari (York University). Register here to partner with them and overcome your AI and policy-related challenges: mila.quebec/en/ai4humani...

3 7

Reposted by Guillaume Lajoie

Mila - Institut québécois d'IA @mila-quebec.bsky.social · Feb 20

Voici deux nouveaux experts du programme Perspectives sur l’IA pour les responsables des politiques : @glajoie.bsky.social (Mila) et Laleh Seyyed-Kalantari (York Univ.) Échangez avec eux et relevez vos défis liés à l'IA et aux politiques. Inscrivez-vous ici mila.quebec/fr/ia-pour-l...

1 3

Reposted by Guillaume Lajoie

The Transmitter @thetransmitter.bsky.social · Feb 7

The earliest studies on necessary and sufficient neural populations were performed on simple invertebrate circuits. In her latest column, @neurograce.bsky.social asks if this logic still serves us as we tackle more sophisticated outputs.

www.thetransmitter.org/systems-neur...

Claims of necessity, sufficiency don’t work well for studies of complex systems

Early studies on necessary and sufficient neural populations were done on simple invertebrate circuits. Does this logic work for complex outputs?

www.thetransmitter.org

1 12 44

Reposted by Guillaume Lajoie

Yoshua Bengio @yoshuabengio.bsky.social · Feb 4

@theguardian.com has produced an excellent recap of some of the key points of the International AI Safety Report. Full article below: www.theguardian.com/technology/2...

What International AI Safety report says on jobs, climate, cyberwar and more

Wide-ranging investigation says impact on work likely to be profound, but opinion on risk of human extinction varies

www.theguardian.com

11 38

Reposted by Guillaume Lajoie

Brokoslaw Laschowski @drlaschowski.bsky.social · Jan 13

Talk by Guillaume Lajoie at the Montreal AI and Neuroscience (MAIN) Conference on credit assignment in neural networks without plasticity.

#neuroscience #neuroAI #AI #compneuro @glajoie.bsky.social www.youtube.com/watch?v=CvCq...

Guillaume Lajoie - Credit assignment in neural networks without plasticity

YouTube video by MAIN Conference

www.youtube.com

5 21