🚨 New preprint alert!
🧠🤖
We propose a theory of how learning curriculum affects generalization through neural population dimensionality. Learning curriculum is a determining factor of neural dimensionality - where you start from determines where you end up.
🧠📈
A 🧵:
tinyurl.com/yr8tawj3
Excited to share that POSSM has been accepted to #NeurIPS2025! See you in San Diego 🏖️
I'm very excited to announce the publication of our new book Neural Interfaces, published by Elsevier. The book is a comprehensive resource for all those interested and gravitating around neural interfaces and brain-computer interfaces (BCIs).
shop.elsevier.com/books/neural...
🐐
Step 1: Understand how scaling improves LLMs.
Step 2: Directly target underlying mechanism.
Step 3: Improve LLMs independent of scale. Profit.
In our ACL 2025 paper we look at Step 1 in terms of training dynamics.
Project: mirandrom.github.io/zsl
Paper: arxiv.org/pdf/2506.05447
(1/n)🚨Train a model solving DFT for any geometry with almost no training data
Introducing Self-Refining Training for Amortized DFT: a variational method that predicts ground-state solutions across geometries and generates its own training data!
📜 arxiv.org/abs/2506.01225
💻 github.com/majhas/self-...
Manitokan are images set up where one can bring a gift or receive a gift. 1930s Rocky Boy Reservation, Montana, Montana State University photograph. Colourized with AI
Preprint Alert 🚀
Multi-agent reinforcement learning (MARL) often assumes that agents know when other agents cooperate with them. But for humans, this isn’t always the case. For example, plains indigenous groups used to leave resources for others to use at effigies called Manitokan.
1/8
Stay tuned for the project page and code, coming soon!
Link: arxiv.org/abs/2506.05320
A big thank you to my co-authors: @nandahkrishna.bsky.social*, @ximengmao.bsky.social*, @mehdiazabou.bsky.social, Eva Dyer, @mattperich.bsky.social, and @glajoie.bsky.social!
🧵7/7
Finally, we show POSSM's performance on speech decoding - a long context task that can quickly grow expensive for Transformers. In the unidirectional setting, POSSM beats the GRU baseline, achieving a phoneme error rate (PER) of 27.3 while having more robustness to variation in preprocessing.
🧵6/7
Cross-species transfer! 🐵➡️🧑
Excitingly, we find that POSSM pretrained solely on monkey reaching data achieves SOTA performance when decoding imagined handwriting in human subjects! This shows the potential of leveraging NHP data to bootstrap human BCI decoding in low-data clinical settings.
🧵5/7
By pretraining on 140 monkey reaching sessions, POSSM effectively transfers to new subjects and tasks, matching or outperforming several baselines (e.g., GRU, POYO, Mamba) across sessions.
✅ High R² across the board
✅ 9× faster inference than Transformers
✅ <5ms latency per prediction
🧵4/7
POSSM combines the real-time inference of an RNN with the tokenization, pretraining, and finetuning abilities of a Transformer!
Using POYO-style tokenization, we encode spikes in 50ms windows and stream them to a recurrent model (e.g., Mamba, GRU) for fast, frequent predictions over time.
🧵3/7
The problem with existing decoders?
😔 RNNs offer efficient, causal inference, but rely on rigid, binned input formats - limiting generalization to new neurons or sessions.
😔 Transformers enable generalization via tokenization, but have high computational costs due to the attention mechanism.
🧵2/7
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
I am joining @ualberta.bsky.social as a faculty member and
@amiithinks.bsky.social!
My research group is recruiting MSc and PhD students at the University of Alberta in Canada. Research topics include generative modeling, representation learning, interpretability, inverse problems, and neuroAI.
Scaling models across multiple animals was a major step toward building neuro-foundation models; the next frontier is enabling multi-task decoding to expand the scope of training data we can leverage.
Excited to share our #ICLR2025 Spotlight paper introducing POYO+ 🧠
poyo-plus.github.io
🧵
Interested in foundation models for #neuroscience? Want to contribute to the development of the next-generation of multi-modal models? Come join us at IVADO in Montreal!
We're hiring a full-time machine learning specialist for this work.
Please share widely!
#NeuroAI 🧠📈 🧪
📽️Recordings from our
@cosynemeeting.bsky.social
#COSYNE2025 workshop on “Agent-Based Models in Neuroscience: Complex Planning, Embodiment, and Beyond" are now online: neuro-agent-models.github.io
🧠🤖
Talk recordings from our COSYNE Workshop on Neuro-foundation Models 🌐🧠 are now up on the workshop website!
neurofm-workshop.github.io
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
Excited to be at #Cosyne2025 for the first time! I'll be presenting my poster [2-104] during the Friday session. E-poster here: www.world-wide.org/cosyne-25/se...
We'll be presenting two projects at #Cosyne2025, representing two main research directions in our lab:
🧠🤖 🧠📈
1/3
@oliviercodol.bsky.social my opportunity to lose to scientists in a different field
Just a couple days until Cosyne - stop by [3-083] this Saturday and say hi! @nandahkrishna.bsky.social
This will be a more difficult Cosyne than normal, due to both the travel restrictions for people coming from the US and the strike that may be happening at the hotel in Montreal.
But, we can still make this an awesome meeting as usual, y'all. Let's pull together and make it happen!
🧠📈
#Cosyne2025
Hi! Currently there are no plans to livestream, but we may *potentially* post recordings in the future (contingent on speaker permission)
Join us at #COSYNE2025 to explore recent advancements in large-scale training and analysis of brain data! 🧠🟦
We also made a starter pack with (most of) our speakers: go.bsky.app/Ss6RaEF
We have a great lineup of speakers and panelists, you can check out our schedule here: neurofm-workshop.github.io. Co-organized with: @mehdiazabou.bsky.social, @nandahkrishna.bsky.social, @colehurwitz.bsky.social, Eva Dyer, and @tyrellturing.bsky.social. We hope to see you there!
How can large-scale models + datasets revolutionize neuroscience 🧠🤖🌐? We are excited to announce our workshop: “Building a foundation model for the brain: datasets, theory, and models” at @cosynemeeting.bsky.social #COSYNE2025. Join us in Mont-Tremblant, Canada from March 31 – April 1!
forms.gle/1DPPVe8KLRWD...
here's a google form for ease!