Lightnews — Scholar-powered news

Shiry Ginosar @shiryginosar.bsky.social · Sep 3

KiVA (Kid-inspired Visual Analogies) Challenge Test Phase is NOW LIVE (Sep 1–Oct 6)!

Can your model reason like a child? Can it beat adults?

🥇 $1,000
🥈 $500 for 2 runner-ups

Join/submit: t.co/zQwA1Nmohy

And join us at @iccv.bsky.social in Hawaii!! 🌴

3 6

Shiry Ginosar @shiryginosar.bsky.social · Jul 18

I am giving a talk this morning at 10:40AM PST as part of the #ICML2025 Workshop on Assessing World Models.

Title: "What Do Vision and Vision-Language Models Really Know About the World?"

Come join us!

www.worldmodelworkshop.org

ICML Workshop on Assessing World Models

Date: Friday, July 18 2025 Time: 8:45am - 5:15pm (Pacific Time) Location: West Ballroom B at ICML 2025 in Vancouver, Canada (Same Floor as Registration)

www.worldmodelworkshop.org

1 1

Reposted by Shiry Ginosar

Dima Damen @dimadamen.bsky.social · Jul 16

Join us for 3rd Perception Test Workshop &Challenge
@iccv.bsky.social #iccv2025
*NEW* this year:
- 3 unified tracks
- novel interpretability track
- guest tracks: KiVA and Physics-IQ
- 4 world-class speakers (see pic)
Up to 50K in prizes sponsored by Google DeepMind
🧵 for details [1/4]

1 3 8

Shiry Ginosar @shiryginosar.bsky.social · Jul 15

Held @iccv.bsky.social in conjunction with GoogleDeepMind 3rd Perception Test Challenge: perception-test-challenge.github.io

Amazing speakers: Ali Farhadi, @alisongopnik.bsky.social @phillipisola.bsky.social, Philipp Krähenbühl.

Fantastically organized by @euniceyiu.bsky.social and co.!

The Third Perception Test Challenge at ICCV 2025

The 3rd Perception Test Challenge at ICCV 2025 evaluates models on physics-aware reasoning, tracking, localization, and VQA with over 50K EUR in prizes.

perception-test-challenge.github.io

1 2

Shiry Ginosar @shiryginosar.bsky.social · Jul 15

🧠How “old” is your model?

Put it to the test with the KiVA Challenge: a new benchmark for abstract visual reasoning, grounded in real developmental data from children and adults.

🏆 Prizes:
🥇$1K to the top model
🥈🥉$500
📅 Deadline: 10/7/25
🔗 kiva-challenge.github.io
@iccv.bsky.social

KiVA Challenge @ ICCV 2025

kiva-challenge.github.io

1 12 22

Shiry Ginosar @shiryginosar.bsky.social · Jun 25

When it comes to goal-directed work, people prioritize controllable variability (a.k.a. empowerment!).

But in undirected play, we shift toward embracing pure variability.

Check out our forthcoming Phil. Trans. A (2026) paper!

1 2

Reposted by Shiry Ginosar

Blake Richards @tyrellturing.bsky.social · Apr 25

Check out our new paper at #ICLR2025, where we show that multi-task neural decoding is both possible and beneficial.

As well, the latents of a model trained only on neural activity capture information about brain regions and cell-types.

Step-by-step, we're gonna scale up folks!

🧠📈 🧪 #NeuroAI

Mehdi Azabou @mehdiazabou.bsky.social · Apr 25

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+

POYO+: Multi-session, multi-task neural decoding from distinct cell-types and brain regions

poyo-plus.github.io

13 53

Reposted by Shiry Ginosar

KQED Forum @kqedforum.bsky.social · Apr 26

🎧 Listen to the podcast!

Professor @alisongopnik.bsky.social and @newamerica.org CEO @slaughteram.bsky.social spoke with @alexis-madrigal.bsky.social about how rethinking our approach to caregiving and how we support care providers could lead to a better society.

🔗:

Alison Gopnik and Anne-Marie Slaughter on Why We’re Not Paying Enough Attention to Caregiving

KQED's Forum · Episode

buff.ly

3 6

Shiry Ginosar @shiryginosar.bsky.social · Apr 23

With Eunice Yiu, Maan Qraitem, Anisa Noor Majhi, Charlie Wong, Yutong Bai, @alisongopnik.bsky.social, and Kate Saenko. @iclr-conf.bsky.social

1 2

Shiry Ginosar @shiryginosar.bsky.social · Apr 23

Think LMMs can reason like a 3-year-old?

Think again!

Our Kid-Inspired Visual Analogies benchmark reveals where young children still win: ey242.github.io/kiva.github....

Catch our #ICLR2025 poster today to see where models still fall short!

Thurs. April 24
3-5:30 pm
Halls 3 + 2B #312

2 7 24

Shiry Ginosar @shiryginosar.bsky.social · Apr 16

Neuroscience is finally taking more and more baby steps towards running experiments at scale!

Cole Hurwitz @colehurwitz.bsky.social · Apr 15

Another step toward a foundation model of the mouse brain: "Neural Encoding and Decoding at Scale (NEDS)"

Trained on neural and behavioral data from 70+ mice, NEDS achieves state-of-the-art prediction of behavior (decoding) and neural responses (encoding) on held-out animals. 🐀

3

Reposted by Shiry Ginosar

The Transmitter @thetransmitter.bsky.social · Apr 15

In his new book, published today, Nachum Ulanovsky calls on the field to embrace naturalistic conditions and move away from overcontrolled experiments.

#neuroskyence

www.thetransmitter.org/systems-neur...

‘Natural Neuroscience,’ an excerpt

In his new book, Nachum Ulanovsky calls on the field to embrace naturalistic conditions and move away from overcontrolled experiments.

www.thetransmitter.org

1 34 81

Shiry Ginosar @shiryginosar.bsky.social · Apr 15

Welcome to TTIC!! We are so excited to have you join us!!

Nick Tomlin @nickatomlin.bsky.social · Apr 15

Writing my first post here to announce that I've accepted an assistant professor job at TTIC! I'll be starting in Fall 2026, and recruiting students this upcoming cycle.

Until then, I'll be wrapping up the PhD at Berkeley, and this summer I'll join NYU as a CDS Faculty Fellow 🏙️

1

Reposted by Shiry Ginosar

carldoersch.bsky.social @carldoersch.bsky.social · Apr 9

We're very excited to introduce TAPNext: a model that sets a new state-of-art for Tracking Any Point in videos, by formulating the task as Next Token Prediction. For more, see: tap-next.github.io

1 9 23

Shiry Ginosar @shiryginosar.bsky.social · Feb 6

This shows the importance of expressive (rather than functional) gesture really nicely in a beautiful non-anthropomorphic robot design

machinelearning.apple.com/research/ele...

ELEGNT: Expressive and Functional Movement Design for Non-Anthropomorphic Robot

Nonverbal behaviors such as posture, gestures, and gaze are essential for conveying internal states, both consciously and unconsciously, in…

machinelearning.apple.com

Reposted by Shiry Ginosar

Noah Snavely @snavely.bsky.social · Feb 6

Did Italo Calvino discover bag of words and topic models in 1979?

She explained to me that a suitably programmed computer can read a novel in a few minutes and record the list of all the words contained in the text, in order of frequency. "That way I can have an already completed reading at hand," Lotaria says, "with an incalculable saving of time. What is the reading of a text, in fact, except the recording of certain thematic recurrences, certain insistences of forms and meanings? An electronic reading supplies me with a list of the frequencies, which I have only to glance at to form an idea of the problems the book suggests to my critical study. Naturally, at the highest frequencies the list records countless articles, pronouns, particles, but I don't pay them any attention. I head straight for the words richest in meaning; they can give me a fairly precise notion of the book."

Lotaria brought me some novels electronically transcribed, in the form of words listed in the order of their frequency. "In a novel of fifty to a hundred thousand words," she said to me, "I advise you to observe immediately the words that are repeated about twenty times. Look here. Words that appear nineteen times:

blood, cartridge belt, commander, do, have, im-
mediately, it, life, seen, sentry, shots, spider, teeth,
together, your...

"Words that appear eighteen times:

boys, cap, come, dead, eat, enough, evening,
French, go, handsome, new, passes, period, po-
tatoes, those, until...

"Don't you already have a clear idea what it's about?" Lotaria says. "There's no question: it's a war novel, all action, brisk writing, with a certain underlying violence. The narration is entirely on the surface, I would say; but

5 2 39

Shiry Ginosar @shiryginosar.bsky.social · Feb 6

That is a fantastic book! But now I realize I don't remember it at all and need to re-read it ;-)

1 1

Reposted by Shiry Ginosar

steven t. piantadosi @spiantado.bsky.social · Feb 5

"That's what the people promulgating these horrible policies want - a bored, indifferent public who figures that who cares, nothing matters any more, it's gonna happen no matter what. But it doesn't have to. Never forget that: it doesn't have to happen."
www.science.org/content/blog...

What's Happening Inside the NIH and NSF

www.science.org

13 21

Shiry Ginosar @shiryginosar.bsky.social · Feb 2

www.youtube.com/watch?v=pVO6...

Classic Sesame Street - What Do People Do All Day

YouTube video by tpirman1982

www.youtube.com

1

Shiry Ginosar @shiryginosar.bsky.social · Feb 2

Sign me up for the project ;-)

1 1

Shiry Ginosar @shiryginosar.bsky.social · Feb 1

What does that mean?
It's both my kids' favorite book ever ;-)

1 1

Reposted by Shiry Ginosar

Riccardo Fusaroli @fusaroli.bsky.social · Jan 19

Why does Western paleolithic cave art strongly prefer animal side views and often abbreviations? Our new #eSymb preprint (osf.io/preprints/ps... w Pagnotta, Psujek, Mendoza Straffon and Tylén ) challenges long-held assumptions about these artistic choices based on cogsci experiments. 1/

5 49 140

Shiry Ginosar @shiryginosar.bsky.social · Jan 9

Fantastic work from @jathushan.bsky.social! With Xinlei Chen, Rulilong Li, Christoph Feichtenhofer, and Jitendra Malik

2

Shiry Ginosar @shiryginosar.bsky.social · Jan 9

New paper! A self-supervised object-centric 2.1D image representation using 3D Gaussians, extending MAE with a Gaussian bottleneck. While Gaussian splatting has been used for single-scene reconstruction, we’re the first to apply it to image representation learning! brjathu.github.io/gmae/.

2 5 37

Shiry Ginosar @shiryginosar.bsky.social · Dec 30

No, but it does a fantastic job of giving feedback if you take it at Harvard in person :-) Does Cornell not have a good undergrad programming class with hands-on feedback to students?

1