Sign in

Avery HW Ryoo

@averyryoo.bsky.social
1.1K followers 301 following 49 posts

i like generative models, science, and Toronto sports teams phd @ mila/udem, prev. @ uwaterloo averyryoo.github.io 🇨🇦🇰🇷

PostsRepliesMedia
Reposted by Avery HW Ryoo
Charlotte Volk @charlottevolk.bsky.social · 30/09/2025
🚨 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
tinyurl.com
The curriculum effect in visual learning: the role of readout dimensionality
Generalization of visual perceptual learning (VPL) to unseen conditions varies across tasks. Previous work suggests that training curriculum may be integral to generalization, yet a theoretical explan...
18326
Reposted by Avery HW Ryoo
Nanda H Krishna @nandahkrishna.bsky.social · 20/09/2025
Excited to share that POSSM has been accepted to #NeurIPS2025! See you in San Diego 🏖️
1113
Reposted by Avery HW Ryoo
Davide Valeriani, PhD 🧠+👨‍💻 @davidevaleriani.bsky.social · 19/08/2025
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...
shop.elsevier.com
Neural Interfaces
Neural Interfaces is a comprehensive book on the foundations, major breakthroughs, and most promising future developments of neural interfaces. The bo
161
Avery HW Ryoo @averyryoo.bsky.social · 12/07/2025
🐐
010
Reposted by Avery HW Ryoo
Andrei Mircea @mirandrom.bsky.social · 12/07/2025
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
241
Reposted by Avery HW Ryoo
majhas.bsky.social @majhas.bsky.social · 10/06/2025
(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-...
1124
Reposted by Avery HW Ryoo
Dane Carnegie Malenfant @dvnxmvlhdf5.bsky.social · 05/06/2025
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
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
13513
Avery HW Ryoo @averyryoo.bsky.social · 06/06/2025
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
arxiv.org
Generalizable, real-time neural decoding with hybrid state-space models
Real-time decoding of neural activity is central to neuroscience and neurotechnology applications, from closed-loop experiments to brain-computer interfaces, where models are subject to strict latency...
061
Avery HW Ryoo @averyryoo.bsky.social · 06/06/2025
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
130
Avery HW Ryoo @averyryoo.bsky.social · 06/06/2025
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
240
Avery HW Ryoo @averyryoo.bsky.social · 06/06/2025
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
130
Avery HW Ryoo @averyryoo.bsky.social · 06/06/2025
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
130
Avery HW Ryoo @averyryoo.bsky.social · 06/06/2025
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
140
Avery HW Ryoo @averyryoo.bsky.social · 06/06/2025
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
25524
Reposted by Avery HW Ryoo
Bahareh Tolooshams @btolooshams.bsky.social · 29/05/2025
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.
1112
Reposted by Avery HW Ryoo
Mehdi Azabou @mehdiazabou.bsky.social · 25/04/2025
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 🧵
poyo-plus.github.io
POYO+
POYO+: Multi-session, multi-task neural decoding from distinct cell-types and brain regions
14510
Reposted by Avery HW Ryoo
Blake Richards @tyrellturing.bsky.social · 11/04/2025
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 🧠📈 🧪
15731
Reposted by Avery HW Ryoo
Satpreet (Sat) Singh @satpreetsingh.bsky.social · 07/04/2025
📽️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 🧠🤖
neuro-agent-models.github.io
🤖 Agent-Based Models in Neuroscience
13711
Reposted by Avery HW Ryoo
Nanda H Krishna @nandahkrishna.bsky.social · 05/04/2025
Talk recordings from our COSYNE Workshop on Neuro-foundation Models 🌐🧠 are now up on the workshop website! neurofm-workshop.github.io
neurofm-workshop.github.io
COSYNE 2025 Workshop - Building a foundation model for the brain
Join us to explore neuro-foundation models. March 31-April 1, 2025 in Mont Tremblant, Canada.
13510
Avery HW Ryoo @averyryoo.bsky.social · 04/04/2025
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
1316
Reposted by Avery HW Ryoo
Charlotte Volk @charlottevolk.bsky.social · 27/03/2025
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...
world-wide.org
How sequential curricula enhance visual learning generalization: The role of subspace dimensionality
COSYNE 2025 e-Poster by Charlotte Volk, Christopher C. Pack, Shahab Bakhtiari
083
Reposted by Avery HW Ryoo
Shahab Bakhtiari @shahabbakht.bsky.social · 27/03/2025
We'll be presenting two projects at #Cosyne2025, representing two main research directions in our lab: 🧠🤖 🧠📈 1/3
1469
Avery HW Ryoo @averyryoo.bsky.social · 27/03/2025
@oliviercodol.bsky.social my opportunity to lose to scientists in a different field
120
Avery HW Ryoo @averyryoo.bsky.social · 24/03/2025
Just a couple days until Cosyne - stop by [3-083] this Saturday and say hi! @nandahkrishna.bsky.social
053
Reposted by Avery HW Ryoo
Blake Richards @tyrellturing.bsky.social · 23/03/2025
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
2445
Avery HW Ryoo @averyryoo.bsky.social · 11/03/2025
Hi! Currently there are no plans to livestream, but we may *potentially* post recordings in the future (contingent on speaker permission)
120
Reposted by Avery HW Ryoo
Mehdi Azabou @mehdiazabou.bsky.social · 10/03/2025
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
go.bsky.app
Neuro-foundation models Workshop - COSYNE 2025
Join the conversation
0174
Avery HW Ryoo @averyryoo.bsky.social · 10/03/2025
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!
neurofm-workshop.github.io
COSYNE 2025 Workshop - Building a foundation model for the brain
Join us to explore neuro-foundation models. March 31-April 1, 2025 in Mont Tremblant, Canada.
150
Avery HW Ryoo @averyryoo.bsky.social · 10/03/2025
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!
24416
Avery HW Ryoo @averyryoo.bsky.social · 01/02/2025
forms.gle/1DPPVe8KLRWD... here's a google form for ease!
forms.gle
000
Avery HW Ryoo @averyryoo.bsky.social · 31/01/2025
Hi! Looking for an undergrad volunteer who's interested in working with SSMs + transformers for neural decoding/BCIs at Mila! Strong coding + Pytorch skills are a must. Please DM/email me your CV + interests (priority given to those based in Montréal). Thanks! 🧠🤖
163
Avery HW Ryoo @averyryoo.bsky.social · 13/01/2025
And don't get me wrong, I think foundation models for neuroscience are yielding fascinating results (currently doing some related work myself!) -- I just have less faith that they'll dominate NeuroAI to the extent that I think they will fields like language, vision, robotics, etc.
010
Avery HW Ryoo @averyryoo.bsky.social · 13/01/2025
Yup! Not to mention the significant investment in computing power, time, and energy that would be required to run and implement (much less fine-tune or train from scratch) these sorts of models, especially for a neuroscience lab that may not have had this type of infra to begin with.
120
Avery HW Ryoo @averyryoo.bsky.social · 13/01/2025
With how broadly/loosely defined NeuroAI is, I'd be pretty cautious about predicting that any one method will dominate the entire field. Maybe I'm being naive, but I find it hard to imagine that methods involving simpler statistical models would invariably be improved by using a huge neural network.
120
Avery HW Ryoo @averyryoo.bsky.social · 28/12/2024
Really excited to read this paper - composing multiple diffusion models/EBMs is something I've been really interested in lately. I think there's potential in this direction for improving the controllability/interpretability of your generation process + mixing and matching pre-trained modules.
030
Avery HW Ryoo @averyryoo.bsky.social · 25/12/2024
Big thanks to my fearless collaborators at @mila-quebec.bsky.social and UdeM: @nandahkrishna.bsky.social, Ximeng Mao, @mattperich.bsky.social, @glajoie.bsky.social
010
Avery HW Ryoo @averyryoo.bsky.social · 25/12/2024
Some exciting news in time for the holidays 🎄🎁☃️ I'll be at Cosyne 2025 (@cosynemeeting.bsky.social) to present our work on generalizable real-time decoding for BCIs 🧠🦾 Really looking forward to seeing everyone in Montréal 🇨🇦! Stay tuned for more details in the new year🤘
1151
Avery HW Ryoo @averyryoo.bsky.social · 25/12/2024
I'm inspired already
010
Avery HW Ryoo @averyryoo.bsky.social · 25/12/2024
When can we expect blog posts?
100
Reposted by Avery HW Ryoo
Tomer Ullman @tomerullman.bsky.social · 20/12/2024
(whipping my around in a crowded party to figure out who just said 'Cocktail Party Effect')
1181
Reposted by Avery HW Ryoo
Dane Carnegie Malenfant @dvnxmvlhdf5.bsky.social · 17/12/2024
Study computer science to avoid writing essays -> Need to write good grant essays to continue studying computer science
161
Reposted by Avery HW Ryoo
Arnaud Doucet @arnauddoucet.bsky.social · 15/12/2024
The slides of my NeurIPS lecture "From Diffusion Models to Schrödinger Bridges - Generative Modeling meets Optimal Transport" can be found here drive.google.com/file/d/1eLa3...
drive.google.com
BreimanLectureNeurIPS2024_Doucet.pdf
932868
Avery HW Ryoo @averyryoo.bsky.social · 12/12/2024
sinthlab EoY social! I'm grateful everyday that I get to work with such a kind and intelligent group of individuals. @mattperich.bsky.social @oliviercodol.bsky.social @anirudhgj.bsky.social
041
Reposted by Avery HW Ryoo
Kirill Neklyudov @k-neklyudov.bsky.social · 10/12/2024
We're presenting our spotlight paper on transition path sampling at #NeurIPS2024 this week! Learn how to speed up the conventional Monte Carlo approaches by orders of magnitude Wed 11 Dec 4:30 pm #2606 arxiv.org/abs/2410.07974 first authors = {Yuanqi Du, Michael Plainer, @brekelmaniac.bsky.social}
0164
Reposted by Avery HW Ryoo
Benno Krojer @bennokrojer.bsky.social · 10/12/2024
Come by tomorrow (Wed) 11am-2pm at @neuripsconf.bsky.social Poster #1606 to chat more about AURORA 🌌, text-guided editing, and why it is arguably more interesting than image generation Or anything related to world models, evals/analysis/interp, vision+language reasoning, cogsci, academic life!
0154
Avery HW Ryoo @averyryoo.bsky.social · 10/12/2024
A big shoutout to my co-authors from the University of Waterloo and the Allen Institute 🇨🇦🇺🇸: Kinjal Patel, @mabuice.bsky.social, Stefan Mihalas, and Bryan Tripp
010
Avery HW Ryoo @averyryoo.bsky.social · 10/12/2024
Thrilled to share a new preprint exploring the spatial organization of multisensory convergence in the mouse isocortex! 🧠🎉 Even more special as it builds on work I started during my undergrad, a lifetime ago 👨‍🦳 Check it out here: www.biorxiv.org/content/10.1...
1134
Avery HW Ryoo @averyryoo.bsky.social · 10/12/2024
Some proof of how much I enjoyed Bixis in the ~7 months I used them this year 😌🚴‍♂️
030
Avery HW Ryoo @averyryoo.bsky.social · 10/12/2024
100% agree w/ Blake here – the biking infrastructure (esp. the amazing Bixi program) might arguably be my fave thing about MTL. But, despite there being a lot of work left to do, I was incredibly heartened to see a big increase of public bike stations the last time I went back home to Toronto 🚲🤞
160
Reposted by Avery HW Ryoo
Yoon @jyoonlee.bsky.social · 07/12/2024
We’re excited to present ET-Flow at #NeurIPS 2024—an Equivariant Flow Matching model that combines simplicity, efficiency, and precision to set a new standard for 3D molecular conformer generation. 🔖Paper: arxiv.org/abs/2410.22388 🔗Github: github.com/shenoynikhil...
1273