I just created a series of seven deep-dive videos about AI, which I've posted to youtube and now here. 😊
Targeted to laypeople, they explore how LLMs work, what they can do, and what impacts they have on learning, well-being, disinformation, the workplace, the economy, and the environment.
Figure shows the methods used in the paper's experiments. In the left column are the methods for Exp 1 (Collection) and in the right are the methods for Exp 2 (Distribution). In each video, three women sit at a table. One sits in the middle, serving as a collector/distributor, and two sit in the foreground with plates. During familiarization trials, resources were collected from or distributed to their plates with an occluder on the screen hiding the outcomes. During test trials, the same videos were played but with the outcomes shown such that infants either viewed an equal collection/distribution or an unequal collection/distribution.
Out in @cognitionjournal.bsky.social with @jaeminhwang.bsky.social, David Sobel (@candmlab.bsky.social), and @jessicas.bsky.social! Most studies of infants’ fairness expectations focus on resource distribution, but in everyday life, we engage in many different kinds of resource exchanges.
Paper title: Language models align with brain regions that represent concepts across modalities. Authors: Maria Ryskina, Greta Tuckute, Alexander Fung, Ashley Malkin, Evelina Fedorenko. Affiliations: Maria is affiliated with the Vector Institute for AI, but the work was done at MIT. All other authors are affiliated with MIT. Email address: maria.ryskina@vectorinstitute.ai.
Interested in language models, brains, and concepts? Check out our COLM 2025 🔦 Spotlight paper!
(And if you’re at COLM, come hear about it on Tuesday – sessions Spotlight 2 & Poster 2)!
I really appreciate the confirmation on the lab logo that this is, indeed, a cat
My Lab at the University of Edinburgh🇬🇧 has funded PhD positions for this cycle!
We study the computational principles of how people learn, reason, and communicate.
It's a new lab, and you will be playing a big role in shaping its culture and foundations.
Spread the words!
Screenshot of the article "How Convincing Is a Crowd? Quantifying the Persuasiveness of a Consensus for Different Individuals and Types of Claims"
We know that a consensus of opinions is persuasive, but how reliable is this effect across people and types of consensus, and are there any kinds of claims where people care less about what other people think? This is what we tested in our new(ish) paper in @psychscience.bsky.social
Link to the paper: escholarship.org/content/qt6x...
A bar chart showing frequencies of answers to the question "Is the mind composed of innate, domain-specialized modules?" with N=197. The most prominent bars are no response (~60), yes (~30) and yo (~30). The rest of the entries are funny, e.g. "I hope so."
Thanks to a rogue Partiful RSVP form at #cogsci2025, I seem to have collected an unexpectedly large dataset (N=197) of whether cognitive scientists think the mind is composed of innate, domain-specialized modules…
CogSci 2025 talk by Rebekah Gelpí: "Resource-rational belief revision can amplify or mitigate polarization"
Good morning #cogsci2025! Join me at Salon 3 today at 14:15 PDT if you're interested in learning how resource-rational models can help us to capture when and why people become polarized.
I don't know how full-featured it is relative to others, but Continue.dev is a nice plugin for VSCode/JetBrains that works with a whole bunch of different LMs. I use it with codestral via local inference right now.
