Lightnews — Scholar-powered news

Reposted by Alex Williams

Melanie Mitchell @melaniemitchell.bsky.social · 1d

Do AI reasoning models abstract and reason like humans?

New paper on this from my group:

arxiv.org/abs/2510.02125

🧵 1/10

Do AI Models Perform Human-like Abstract Reasoning Across Modalities?

OpenAI's o3-preview reasoning model exceeded human accuracy on the ARC-AGI benchmark, but does that mean state-of-the-art models recognize and reason with the abstractions that the task creators inten...

arxiv.org

3 22 79

Reposted by Alex Williams

Science and Freedom Alliance - SAFA @safa-science.bsky.social · 6d

Federal workers are speaking out on how the Trump regime is harming Americans:

"We stand here today at a moment of the greatest assault of our lifetimes on our freedoms. The US Constitution is being shredded... We are here to call on Congress to stand up and stop it."

youtu.be/uw1KwIBgrCw?...

Mark Histed, NIH Scientist *speaking in personal capacity

YouTube video by Civil Servants Coalition

youtu.be

6 170 340

Reposted by Alex Williams

Jonathan A. Michaels @jonathanamichaels.bsky.social · 6d

The neural control & computation lab is recruiting!

If you're interested in using large-scale neural population recordings to study how the brain learns to produce complex and flexible behaviours, please get in touch.

www.ncclab.ca

5 43 95

Reposted by Alex Williams

Michael Kirchhof (ICML) @mkirchhof.bsky.social · 7d

Many treat uncertainty = a number. At Apple, we're rethinking this: LLMs should output strings that reveal all information of their internal distributions. We find that Reasoning, SFT, CoT can't do it - yet. To get there, we introduce the SelfReflect benchmark.

arxiv.org/pdf/2505.20295

3 6 30

Reposted by Alex Williams

Fanny Cazettes @fannycazettes.bsky.social · 8d

Did you know that facial expressions reveal more than meets the eye? 🤯

Our new study shows that even a mouse's face 🐭 can reflect hidden neural computations🧠. Turns out, facial expressions are more than just emotions!

We're so excited to see this paper out @natneuro.nature.com 🎉
🔗: rdcu.be/eIQzO

Facial expressions in mice reveal latent cognitive variables and their neural correlates

Nature Neuroscience - The face reveals more than just emotion. Cazettes, Reato and colleagues show that subtle facial movements reveal hidden cognitive states, reflecting the brain’s ongoing...

rdcu.be

2 13 60

Reposted by Alex Williams

Laureline Logiaco @laurelinelogiaco.bsky.social · 10d

Interested in doing a Ph.D. to work on building models of the brain/behavior? Consider applying to graduate schools at CU Anschutz:
1. Neuroscience www.cuanschutz.edu/graduate-pro...
2. Bioengineering engineering.ucdenver.edu/bioengineeri...

You could work with several comp neuro PIs, including me.

1 30 51

Reposted by Alex Williams

Matt Golub @mattgolub.bsky.social · 12d

(1/n) How can we infer single-trial communication between 🧠 regions? Check out our ICML 2025 paper: “Accurate Identification of Communication Between Multiple Interacting Neural Populations,” by Belle Liu and @jsacks.bsky.social in my lab: arxiv.org/abs/2506.19094

Accurate identification of communication between multiple interacting neural populations

Neural recording technologies now enable simultaneous recording of population activity across many brain regions, motivating the development of data-driven models of communication between brain region...

arxiv.org

1 8 24

Alex Williams @itsneuronal.bsky.social · 11d

This is the way

bsky.app/profile/itsn...

Alex Williams @itsneuronal.bsky.social · Jul 31

Easy two part solution.

(a) NIH should run its own journal, in house, that pays reviewers for their time and waives all publication fees for any NIH funded project.

(b) Publication fees for any other journal are then not an allowable budget line on NIH grants.

Harmit Singh Malik @harmitmalik.bsky.social · Jul 31

NIH wants to hear from you. No really.

Request for Information on Maximizing Research Funds by Limiting Allowable Publishing Costs
grants.nih.gov/grants/guide...

1

Reposted by Alex Williams

Cian O'Donnell @cianodonnell.bsky.social · 11d

I still think the major funders should all set up their own PLoS One-style journals (peer reviewed for rigour but not impact) and mandate that if you take their money, you publish the resulting research there, full stop

Richard Sever @richardsever.bsky.social · 13d

HHMI adopts Plan U journals.plos.org/plosbiology/...

4 2 9

Reposted by Alex Williams

Jason Williams @jasonwilliamsny.bsky.social · 12d

Please share with anyone who cares about NSF support for graduate students and take 30 seconds to sign and leave a comment.

The deadline for the 2025 Graduate Research Fellowship Program is about one month away and literally no one can apply. #NSFGRFP

jasonjwilliamsny.github.io/grfp2025/

An Open Letter to U.S. STEM Leadership on the NSF Graduate Research Fellowship Program

jasonjwilliamsny.github.io

12 270 220

Reposted by Alex Williams

Kris Jensen @kristorpjensen.bsky.social · 14d

I’m super excited to finally put my recent work with @behrenstimb.bsky.social on bioRxiv, where we develop a new mechanistic theory of how PFC structures adaptive behaviour using attractor dynamics in space and time!

www.biorxiv.org/content/10.1...

9 85 200

Reposted by Alex Williams

Richard Gao @rdgao.bsky.social · 15d

I've been waiting some years to make this joke and now it’s real:

I conned somebody into giving me a faculty job!

I’m starting as a W1 Tenure-Track Professor at Goethe University Frankfurt in a week (lol), in the Faculty of CS and Math

and I'm recruiting PhD students 🤗

a man wearing a white shirt and tie smiles in front of a window

ALT: a man wearing a white shirt and tie smiles in front of a window

media.tenor.com

30 31 180

Reposted by Alex Williams

Marlene Cohen @marlenecohen.bsky.social · 15d

New preprint! How can you remember an image you saw once, even after seeing thousands of them? We find a role for humble mid-level visual cortex in high-capacity, one-shot learning. doi.org/10.1101/2025.09.22.677855 🧵🧪1/

Neuronal signatures of successful one-shot memory in mid-level visual cortex

High-capacity, one-shot visual recognition memory challenges theories of learning and neural coding because it requires rapid, robust, and durable representations. Most studies have focused on the hip...

doi.org

3 28 89

Reposted by Alex Williams

Angie Rasmussen @angierasmussen.bsky.social · 16d

Schulman thinks Jay isn’t a hardened cynic who is carrying out MAHA marching orders, but an idealist who has naively stumbled into being used as a tool of authoritarianism. He’s just too gosh darn nice & devoted to reason to do this willingly. This view is far too charitable.

2 5 74

Reposted by Alex Williams

Carl Zimmer @carlzimmer.com · 24d

A deep dive into the destruction of US cancer research by @jonathanmahler.bsky.social “It’s an absolutely unmitigated disaster,” a former top official at NIH told him. “It will take decades to recover from this, if we ever do.” Gift link: nyti.ms/48iH3Cr

nyti.ms

12 220 400

Reposted by Alex Williams

Gregg Gonsalves @gregggonsalves.bsky.social · 25d

Though it’s traumatic for me to bring this up, I was targeted by a group allied with Turning Point USA in 2018, with the help of Tucker Carlson on Fox. My home address circulated online and I got death threats. 1/

92 1.1K 3K

Reposted by Alex Williams

Jonathan Pillow @jpillowtime.bsky.social · 26d

Great opportunity to learn to use fancy neural data analysis tools developed at the @flatironinstitute.org. Sign up for this workshop at SFN 2025!

Pynapple @pynapple.bsky.social · 26d

🎉 Deadline is *tomorrow*!
Join the @flatironinstitute.org workshop on pynapple & NeMos @sfn.org this November to analyze and model neural data

pynapple: analyze neural time series
NeMoS: model neural population dynamics
Meals + accommodation included
👉 neurorse.flatironinstitute.org/events/2025/...

SfN 2025

neurorse.flatironinstitute.org

4 10

Reposted by Alex Williams

Pynapple @pynapple.bsky.social · 26d

🎉 Deadline is *tomorrow*!
Join the @flatironinstitute.org workshop on pynapple & NeMos @sfn.org this November to analyze and model neural data

pynapple: analyze neural time series
NeMoS: model neural population dynamics
Meals + accommodation included
👉 neurorse.flatironinstitute.org/events/2025/...

SfN 2025

neurorse.flatironinstitute.org

7 5

Reposted by Alex Williams

Pynapple @pynapple.bsky.social · Aug 25

New workshop! Just before SFN, join the Flatiron Institute Center for Computational Neuroscience of the Simons Foundation for a workshop on pynapple and NeMoS. Learn how to use these open-source packages to analyze and model neural data! Accommodation & meals provided. Link ⬇️

1 9 14

Reposted by Alex Williams

Ricard Solé @ricardsole.bsky.social · 28d

Can a single cell learn? Even without a brain, some microbes show simple forms of cognition. Can this basal cognition be engineered? Check our new paper with @jordiplam.bsky.social on the minimal synthetic circuits & their cognitive limits. @drmichaellevin.bsky.social www.biorxiv.org/content/10.1...

4 42 110

Reposted by Alex Williams

Jason Shepherd @jasonsynaptic.bsky.social · 27d

We are not safe on Utah education campuses. State laws allow open carry guns and our politicians fan the flames of hate and division. The USA will never be a “free” country until something is actually done about gun violence! I fear for where we are heading 😢

3 37

Reposted by Alex Williams

Pynapple @pynapple.bsky.social · 29d

⏳ Deadline is next Friday! Apply now to join our #SfN25 workshop on pynapple & nemos. Learn to analyze & model neural data using software developed at @flatironinstitute.org
Accommodations + meals included! 👉

SfN 2025

neurorse.flatironinstitute.org