Simon Schug
@smonsays.bsky.social
📤 940
📥 238
📝 22
postdoc @princeton computational cognitive science ∪ machine learning
https://smn.one
pinned post!
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.
over 1 year ago
1
11
0
LLM agents are a serious problem for online experiments. It is very easy to use them and very hard to spot them. What can researchers do? With
@brendenlake.bsky.social
, we suggest detecting LLMs based on their lack of human cognitive constraints in our
#CogSci2026
paper:
arxiv.org/abs/2604.00016
loading . . .
19 days ago
0
24
5
reposted by
Simon Schug
Minas Karamanis
29 days ago
Hey, I wrote a thing about AI in astrophysics
ergosphere.blog/posts/the-ma...
loading . . .
The machines are fine. I'm worried about us.
On AI agents, grunt work, and the part of science that isn't replaceable.
https://ergosphere.blog/posts/the-machines-are-fine/
109
1729
785
reposted by
Simon Schug
Mark Histed
about 1 month ago
The hardest part of science is posing the right question, not answering it.
add a skeleton here at some point
4
242
54
reposted by
Simon Schug
Kai Sandbrink
about 1 month ago
Excited that my paper on metacontrol in humans and neural networks with
@summerfieldlab.bsky.social
and
@lhuntneuro.bsky.social
is out in PNAS! We examine the way that predictive representations of control enable behavioral adaptation across settings, and pathologies:
www.pnas.org/doi/10.1073/...
0
38
16
reposted by
Simon Schug
Sam Gershman
about 2 months ago
I think Fodor & Pylyshyn's 1988 paper is possibly the most mischaracterized paper in the history of cognitive science. It's often cited as arguing that neural networks cannot achieve systematicity, compositionality, and productivity. But that's not what they actually argue...
2
90
22
reposted by
Simon Schug
Jonathan Nicholas
3 months ago
Our experiences have countless details, and it can be hard to know which matter. How can we behave effectively in the future when, right now, we don't know what we'll need? Out today in
@nathumbehav.nature.com
,
@marcelomattar.bsky.social
and I find that people solve this by using episodic memory.
loading . . .
Episodic memory facilitates flexible decision-making via access to detailed events - Nature Human Behaviour
Nicholas and Mattar found that people use episodic memory to make decisions when it is unclear what will be needed in the future. These findings reveal how the rich representational capacity of episod...
https://www.nature.com/articles/s41562-025-02383-3
7
130
51
reposted by
Simon Schug
C. Thi Nguyen
3 months ago
Last term I tried an experiment: I walked into my Tech and Design Ethics class, admitted that I had *no idea* what to do about ChatGPT - so I would let them figure it out. As in: their first project was to decide and write the ChatGPT policy for the class. Here's what happened:
25
2363
1105
reposted by
Simon Schug
Griffiths Computational Cognitive Science Lab
4 months ago
Excited to announce a new book telling the story of mathematical approaches to studying the mind, from the origins of cognitive science to modern AI! The Laws of Thought will be published in February and is available for pre-order now.
2
167
46
reposted by
Simon Schug
Erin Grant
5 months ago
Thrilled to start 2026 as faculty in Psych & CS
@ualberta.bsky.social
+
Amii.ca
Fellow! 🥳 Recruiting students to develop theories of cognition in natural & artificial systems 🤖💭🧠. Find me at
#NeurIPS2025
workshops (speaking
coginterp.github.io/neurips2025
& organising
@dataonbrainmind.bsky.social
)
4
105
28
reposted by
Simon Schug
Paul Masset
5 months ago
I am recruiting graduate students for the experimental side of my lab
@mcgill.ca
for admission in Fall 2026! Get in touch if you're interested in how brain circuits implement distributed computation, including dopamine-based distributed RL and probabilistic representations.
1
37
30
reposted by
Simon Schug
Brenden Lake
6 months ago
Checking out the Princeton trails on our lab retreat
0
28
1
Does scaling lead to compositional generaliztation? Our
#NeurIPS2025
Spotlight paper suggests that it can -- with the right training distribution. 🧵 A short thread:
6 months ago
1
15
1
reposted by
Simon Schug
Brenden Lake
11 months ago
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!
4
47
2
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...
about 1 year ago
1
10
0
reposted by
Simon Schug
Taylor Webb
about 1 year ago
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)
loading . . .
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 ...
https://arxiv.org/abs/2502.20332
5
116
37
reposted by
Simon Schug
Konrad Kording
about 1 year ago
New blog post: The principle of neuroscience.
medium.com/@kording/the...
loading . . .
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…
https://medium.com/@kording/the-principle-of-neural-science-92eb016c95e8
3
46
6
reposted by
Simon Schug
Marine Schimel
about 1 year ago
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...
loading . . .
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...
https://www.biorxiv.org/content/10.1101/2024.11.27.625691v1
1
19
12
reposted by
Simon Schug
Guillaume Bellec
over 1 year ago
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
4
125
51
reposted by
Simon Schug
Mark D Humphries
over 1 year ago
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...
loading . . .
2024: A Review of the Year in Neuroscience
Feeling a bit wired
https://medium.com/the-spike/2024-a-review-of-the-year-in-neuroscience-84d343155146
7
190
90
reposted by
Simon Schug
Kris Jensen
over 1 year ago
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!
loading . . .
An introduction to reinforcement learning for neuroscience | Published in Neurons, Behavior, Data analysis, and Theory
By Kristopher T. Jensen. Reinforcement learning for neuroscientists
https://tinyurl.com/5f58zdy3
6
129
32
reposted by
Simon Schug
Ben Recht
over 1 year ago
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...
loading . . .
Monster Models
Systems-level biology is hard because systems-level engineering is hard.
https://www.argmin.net/p/monster-models
3
12
3
reposted by
Simon Schug
Badr AlKhamissi
over 1 year ago
🚨 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
🧵👇
2
103
31
reposted by
Simon Schug
Blake Richards
over 1 year ago
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
add a skeleton here at some point
21
152
55
reposted by
Simon Schug
Razvan Pascanu
over 1 year ago
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 ..
add a skeleton here at some point
1
32
7
reposted by
Simon Schug
Markus Meister
over 1 year ago
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?
1
39
12
reposted by
Simon Schug
Andrew Lampinen
over 1 year ago
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
loading . . .
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...
https://arxiv.org/abs/2412.03782
2
122
33
reposted by
Simon Schug
Kai Sandbrink
over 1 year ago
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.
2
65
16
reposted by
Simon Schug
Blake Richards
over 1 year ago
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. 🧠📈 🧪
add a skeleton here at some point
4
68
10
reposted by
Simon Schug
Griffiths Computational Cognitive Science Lab
over 1 year ago
(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...
15
521
135
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
add a skeleton here at some point
over 1 year ago
33
50
18
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.
over 1 year ago
1
11
0
you reached the end!!
feeds!
log in