Simon Schug
@smonsays.bsky.social
Pinned
Simon Schug
@smonsays.bsky.social
· Oct 28
Neural networks used to struggle with compositionality but transformers got really good at it. How come?
And why does attention work so much better with multiple heads?
There might be a common answer to both of these questions.
And why does attention work so much better with multiple heads?
There might be a common answer to both of these questions.
Reposted by Simon Schug
Checking out the Princeton trails on our lab retreat
November 3, 2025 at 5:44 PM
Checking out the Princeton trails on our lab retreat
Does scaling lead to compositional generaliztation?
Our #NeurIPS2025 Spotlight paper suggests that it can -- with the right training distribution.
🧵 A short thread:
Our #NeurIPS2025 Spotlight paper suggests that it can -- with the right training distribution.
🧵 A short thread:
November 4, 2025 at 2:33 PM
Does scaling lead to compositional generaliztation?
Our #NeurIPS2025 Spotlight paper suggests that it can -- with the right training distribution.
🧵 A short thread:
Our #NeurIPS2025 Spotlight paper suggests that it can -- with the right training distribution.
🧵 A short thread:
Reposted by Simon Schug
I'm joining Princeton University as an Associate Professor of Computer Science and Psychology this fall! Princeton is ambitiously investing in AI and Natural & Artificial Minds, and I'm excited for my lab to contribute. Recruiting postdocs and Ph.D. students in CS and Psychology — join us!
June 12, 2025 at 2:29 PM
I'm joining Princeton University as an Associate Professor of Computer Science and Psychology this fall! Princeton is ambitiously investing in AI and Natural & Artificial Minds, and I'm excited for my lab to contribute. Recruiting postdocs and Ph.D. students in CS and Psychology — join us!
Are transformers smarter than you? Hypernetworks might explain why.
Come checkout our Oral at #ICLR tomorrow (Apr 26th, poster at 10:00, Oral session 6C in the afternoon).
openreview.net/forum?id=V4K...
Come checkout our Oral at #ICLR tomorrow (Apr 26th, poster at 10:00, Oral session 6C in the afternoon).
openreview.net/forum?id=V4K...
April 25, 2025 at 4:50 AM
Are transformers smarter than you? Hypernetworks might explain why.
Come checkout our Oral at #ICLR tomorrow (Apr 26th, poster at 10:00, Oral session 6C in the afternoon).
openreview.net/forum?id=V4K...
Come checkout our Oral at #ICLR tomorrow (Apr 26th, poster at 10:00, Oral session 6C in the afternoon).
openreview.net/forum?id=V4K...
Reposted by Simon Schug
LLMs have shown impressive performance in some reasoning tasks, but what internal mechanisms do they use to solve these tasks? In a new preprint, we find evidence that abstract reasoning in LLMs depends on an emergent form of symbol processing arxiv.org/abs/2502.20332 (1/N)
Emergent Symbolic Mechanisms Support Abstract Reasoning in Large Language Models
Many recent studies have found evidence for emergent reasoning capabilities in large language models, but debate persists concerning the robustness of these capabilities, and the extent to which they ...
arxiv.org
March 10, 2025 at 7:08 PM
LLMs have shown impressive performance in some reasoning tasks, but what internal mechanisms do they use to solve these tasks? In a new preprint, we find evidence that abstract reasoning in LLMs depends on an emergent form of symbol processing arxiv.org/abs/2502.20332 (1/N)
Reposted by Simon Schug
New blog post: The principle of neuroscience. medium.com/@kording/the...
The Principle of Neural Science
I first encountered Principles of Neural Science as a young student of neuroscience. The book was filled with delightful narratives…
medium.com
February 14, 2025 at 11:04 PM
New blog post: The principle of neuroscience. medium.com/@kording/the...
Reposted by Simon Schug
For my first Bluesky post, I'm very excited to share a thread on our recent work with Mitra Javadzadeh, investigating how connections between cortical areas shape computations in the neocortex! [1/7] www.biorxiv.org/content/10.1...
Dynamic consensus-building between neocortical areas via long-range connections
The neocortex is organized into functionally specialized areas. While the functions and underlying neural circuitry of individual neocortical areas are well studied, it is unclear how these regions op...
www.biorxiv.org
January 31, 2025 at 2:57 AM
For my first Bluesky post, I'm very excited to share a thread on our recent work with Mitra Javadzadeh, investigating how connections between cortical areas shape computations in the neocortex! [1/7] www.biorxiv.org/content/10.1...
Reposted by Simon Schug
Pre-print 🧠🧪
Is mechanism modeling dead in the AI era?
ML models trained to predict neural activity fail to generalize to unseen opto perturbations. But mechanism modeling can solve that.
We say "perturbation testing" is the right way to evaluate mechanisms in data-constrained models
1/8
Is mechanism modeling dead in the AI era?
ML models trained to predict neural activity fail to generalize to unseen opto perturbations. But mechanism modeling can solve that.
We say "perturbation testing" is the right way to evaluate mechanisms in data-constrained models
1/8
January 8, 2025 at 4:33 PM
Pre-print 🧠🧪
Is mechanism modeling dead in the AI era?
ML models trained to predict neural activity fail to generalize to unseen opto perturbations. But mechanism modeling can solve that.
We say "perturbation testing" is the right way to evaluate mechanisms in data-constrained models
1/8
Is mechanism modeling dead in the AI era?
ML models trained to predict neural activity fail to generalize to unseen opto perturbations. But mechanism modeling can solve that.
We say "perturbation testing" is the right way to evaluate mechanisms in data-constrained models
1/8
Reposted by Simon Schug
Cutting it a bit fine, but here’s my review of the year in neuroscience for 2024
The eighth of these, would you believe? We’ve got dark neurons, tiny monkeys, the most complete brain wiring diagram ever constructed, and much more…
Published on The Spike
Enjoy!
medium.com/the-spike/20...
The eighth of these, would you believe? We’ve got dark neurons, tiny monkeys, the most complete brain wiring diagram ever constructed, and much more…
Published on The Spike
Enjoy!
medium.com/the-spike/20...
2024: A Review of the Year in Neuroscience
Feeling a bit wired
medium.com
December 30, 2024 at 4:00 PM
Cutting it a bit fine, but here’s my review of the year in neuroscience for 2024
The eighth of these, would you believe? We’ve got dark neurons, tiny monkeys, the most complete brain wiring diagram ever constructed, and much more…
Published on The Spike
Enjoy!
medium.com/the-spike/20...
The eighth of these, would you believe? We’ve got dark neurons, tiny monkeys, the most complete brain wiring diagram ever constructed, and much more…
Published on The Spike
Enjoy!
medium.com/the-spike/20...
Reposted by Simon Schug
I wrote an introduction to RL for neuroscience last year that was just published in NBDT: tinyurl.com/5f58zdy3
This review aims to provide some intuition for and derivations of RL methods commonly used in systems neuroscience, ranging from TD learning through the SR to deep and distributional RL!
This review aims to provide some intuition for and derivations of RL methods commonly used in systems neuroscience, ranging from TD learning through the SR to deep and distributional RL!
An introduction to reinforcement learning for neuroscience | Published in Neurons, Behavior, Data analysis, and Theory
By Kristopher T. Jensen. Reinforcement learning for neuroscientists
tinyurl.com
December 21, 2024 at 5:59 PM
I wrote an introduction to RL for neuroscience last year that was just published in NBDT: tinyurl.com/5f58zdy3
This review aims to provide some intuition for and derivations of RL methods commonly used in systems neuroscience, ranging from TD learning through the SR to deep and distributional RL!
This review aims to provide some intuition for and derivations of RL methods commonly used in systems neuroscience, ranging from TD learning through the SR to deep and distributional RL!
Reposted by Simon Schug
Stitching component models into system models has proven difficult in biology. But how much easier has it been in engineering? www.argmin.net/p/monster-mo...
Monster Models
Systems-level biology is hard because systems-level engineering is hard.
www.argmin.net
December 20, 2024 at 3:29 PM
Stitching component models into system models has proven difficult in biology. But how much easier has it been in engineering? www.argmin.net/p/monster-mo...
Reposted by Simon Schug
🚨 New Paper!
Can neuroscience localizers uncover brain-like functional specializations in LLMs? 🧠🤖
Yes! We analyzed 18 LLMs and found units mirroring the brain's language, theory of mind, and multiple demand networks!
w/ @gretatuckute.bsky.social, @abosselut.bsky.social, @mschrimpf.bsky.social
🧵👇
Can neuroscience localizers uncover brain-like functional specializations in LLMs? 🧠🤖
Yes! We analyzed 18 LLMs and found units mirroring the brain's language, theory of mind, and multiple demand networks!
w/ @gretatuckute.bsky.social, @abosselut.bsky.social, @mschrimpf.bsky.social
🧵👇
December 19, 2024 at 3:06 PM
🚨 New Paper!
Can neuroscience localizers uncover brain-like functional specializations in LLMs? 🧠🤖
Yes! We analyzed 18 LLMs and found units mirroring the brain's language, theory of mind, and multiple demand networks!
w/ @gretatuckute.bsky.social, @abosselut.bsky.social, @mschrimpf.bsky.social
🧵👇
Can neuroscience localizers uncover brain-like functional specializations in LLMs? 🧠🤖
Yes! We analyzed 18 LLMs and found units mirroring the brain's language, theory of mind, and multiple demand networks!
w/ @gretatuckute.bsky.social, @abosselut.bsky.social, @mschrimpf.bsky.social
🧵👇
Reposted by Simon Schug
1/ Okay, one thing that has been revealed to me from the replies to this is that many people don't know (or refuse to recognize) the following fact:
The unts in ANN are actually not a terrible approximation of how real neurons work!
A tiny 🧵.
🧠📈 #NeuroAI #MLSky
The unts in ANN are actually not a terrible approximation of how real neurons work!
A tiny 🧵.
🧠📈 #NeuroAI #MLSky
Why does anyone have any issue with this?
I've seen people suggesting it's problematic, that neuroscientists won't like it, and so on.
But, I literally don't see why this is problematic...
I've seen people suggesting it's problematic, that neuroscientists won't like it, and so on.
But, I literally don't see why this is problematic...
This would be funny if it weren't sad...
Coming from the "giants" of AI.
Or maybe this was posted out of context? Please clarify.
I can't process this...
Coming from the "giants" of AI.
Or maybe this was posted out of context? Please clarify.
I can't process this...
December 16, 2024 at 8:03 PM
Reposted by Simon Schug
For my first post on Bluesky .. I'll start by announcing our 2025 edition of EEML which will be in Sarajevo :) ! I'm really excited about it and hope to see many of you there. Please follow the website (and Bluesky account) for more details which are coming soon ..
Hello Bluesky! 🦋
This will be the official account of the Eastern European Machine Learning (EEML) community.
Follow us for news regarding our summer schools, workshops, education/community initiatives, and more!
This will be the official account of the Eastern European Machine Learning (EEML) community.
Follow us for news regarding our summer schools, workshops, education/community initiatives, and more!
December 15, 2024 at 6:39 PM
For my first post on Bluesky .. I'll start by announcing our 2025 edition of EEML which will be in Sarajevo :) ! I'm really excited about it and hope to see many of you there. Please follow the website (and Bluesky account) for more details which are coming soon ..
Reposted by Simon Schug
Have you had private doubts whether we'll ever understand the brain? Whether we'll be able explain psychological phenomena in an exhaustive way that ranges from molecules to membranes to synapses to cells to cell types to circuits to computation to perception and behavior?
November 14, 2024 at 5:18 AM
Have you had private doubts whether we'll ever understand the brain? Whether we'll be able explain psychological phenomena in an exhaustive way that ranges from molecules to membranes to synapses to cells to cell types to circuits to computation to perception and behavior?
Reposted by Simon Schug
What counts as in-context learning (ICL)? Typically, you might think of it as learning a task from a few examples. However, we’ve just written a perspective (arxiv.org/abs/2412.03782) suggesting interpreting a much broader spectrum of behaviors as ICL! Quick summary thread: 1/7
The broader spectrum of in-context learning
The ability of language models to learn a task from a few examples in context has generated substantial interest. Here, we provide a perspective that situates this type of supervised few-shot learning...
arxiv.org
December 10, 2024 at 6:17 PM
What counts as in-context learning (ICL)? Typically, you might think of it as learning a task from a few examples. However, we’ve just written a perspective (arxiv.org/abs/2412.03782) suggesting interpreting a much broader spectrum of behaviors as ICL! Quick summary thread: 1/7
Reposted by Simon Schug
Thrilled to share our NeurIPS Spotlight paper with Jan Bauer*, @aproca.bsky.social*, @saxelab.bsky.social, @summerfieldlab.bsky.social, Ali Hummos*! openreview.net/pdf?id=AbTpJ...
We study how task abstractions emerge in gated linear networks and how they support cognitive flexibility.
We study how task abstractions emerge in gated linear networks and how they support cognitive flexibility.
December 3, 2024 at 4:05 PM
Thrilled to share our NeurIPS Spotlight paper with Jan Bauer*, @aproca.bsky.social*, @saxelab.bsky.social, @summerfieldlab.bsky.social, Ali Hummos*! openreview.net/pdf?id=AbTpJ...
We study how task abstractions emerge in gated linear networks and how they support cognitive flexibility.
We study how task abstractions emerge in gated linear networks and how they support cognitive flexibility.
Reposted by Simon Schug
Great thread from @michaelhendricks.bsky.social!
Reminds me of something Larry Abbott once said to me at a summer school:
Many physicists come into neuroscience assuming that the failure to find laws of the brain was just because biologists aren't clever enough. In fact, there are no laws.
🧠📈 🧪
Reminds me of something Larry Abbott once said to me at a summer school:
Many physicists come into neuroscience assuming that the failure to find laws of the brain was just because biologists aren't clever enough. In fact, there are no laws.
🧠📈 🧪
I came across a quote in an article, which I will paraphrase: the ultimate goal of neuroscience is to model the brain and derive laws that define the brain’s computational abilities. Statements like this are common and presented as self-evident, but I think they are wrong.
November 13, 2024 at 6:49 PM
Great thread from @michaelhendricks.bsky.social!
Reminds me of something Larry Abbott once said to me at a summer school:
Many physicists come into neuroscience assuming that the failure to find laws of the brain was just because biologists aren't clever enough. In fact, there are no laws.
🧠📈 🧪
Reminds me of something Larry Abbott once said to me at a summer school:
Many physicists come into neuroscience assuming that the failure to find laws of the brain was just because biologists aren't clever enough. In fact, there are no laws.
🧠📈 🧪
Reposted by Simon Schug
(1/5) Very excited to announce the publication of Bayesian Models of Cognition: Reverse Engineering the Mind. More than a decade in the making, it's a big (600+ pages) beautiful book covering both the basics and recent work: mitpress.mit.edu/978026204941...
November 19, 2024 at 9:33 PM
(1/5) Very excited to announce the publication of Bayesian Models of Cognition: Reverse Engineering the Mind. More than a decade in the making, it's a big (600+ pages) beautiful book covering both the basics and recent work: mitpress.mit.edu/978026204941...
To help find people at the intersection of neuroscience and AI. Of course let me know if I missed someone or you’d like to be added 🧪 🧠
#neuroskyence
go.bsky.app/CAfmKQs
#neuroskyence
go.bsky.app/CAfmKQs
November 13, 2024 at 3:26 PM
To help find people at the intersection of neuroscience and AI. Of course let me know if I missed someone or you’d like to be added 🧪 🧠
#neuroskyence
go.bsky.app/CAfmKQs
#neuroskyence
go.bsky.app/CAfmKQs
Neural networks used to struggle with compositionality but transformers got really good at it. How come?
And why does attention work so much better with multiple heads?
There might be a common answer to both of these questions.
And why does attention work so much better with multiple heads?
There might be a common answer to both of these questions.
October 28, 2024 at 3:22 PM
Neural networks used to struggle with compositionality but transformers got really good at it. How come?
And why does attention work so much better with multiple heads?
There might be a common answer to both of these questions.
And why does attention work so much better with multiple heads?
There might be a common answer to both of these questions.