Max Puelma Touzel

Social Simulation with LLMs Important Information Workshop Date: October 10, 2025 Workshop Room: 523AB Workshop Time Zone: Montreal (UTC -4) Contact Email: social-simulation@googlegroups.com

Excited to be at #COLM2025 and to be organizing one of (perhaps the?) first workshop on social modelling with LLMs! Looking forward to this community-wide stocktake on the state of the field, main challenges, and promising applications!
sites.google.com/view/social-...

Stochastic Thermodynamics Cambridge Core - Statistical Physics - Stochastic Thermodynamics

Looks like I missed the publication of another stochastic thermodynamics text in June. Seifert is arguably better positioned, but let's see about his exposition style. The other text is Politi & Pigolotti, which was nice, but light on the info side.
www.cambridge.org/core/books/s...

YouTube video by Movieclips 24 Hour Party People (2002) - Tony Doesn't Sell Out Scene (10/12) | Movieclips

Another model is to project a pragmatic view that attracts more options, but build in structure to your incentives to keep you from selling out. Tony Wilson shows us the way: www.youtube.com/watch?v=fwkz...

trying to parse this. What does the "without focusing on intervening on particular nodes." mean? You mean other system level interventions other than structure?

🚨Our preprint is online!🚨

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

How do #dopamine neurons perform the key calculations in reinforcement #learning?

Read on to find out more! 🧵

This is it

I am recruiting (Canadian🇨🇦) graduate students at York University for Fall 2026 in Social and Personality Psychology! If you are interested in misinformation, political polarization, and/or computational social science, I'd love to hear from you. Feel free to reach out with any questions.

The counterview I find compelling is optimality as a crucial feature of methodology: you carefully pick as objects of study behaviours that are very likely to have strong selection pressure, thereby taking on forms that you can produce as solutions. How else could you make predictions?

First page of the paper. Title: "Misinformation as a Harm: Structured Approaches for Fact-Checking Prioritization"

It seems obvious that we should prioritize addressing misinformation that is more harmful. But what makes something more likely to be harmful? Can we can reason about it before the harm has occurred?

📣 In an upcoming #CSCW2024 paper, we present a taxonomy of *Misinformation as a Harm* ➡️

Similar experience. I moved into more ML where I saw (at least locally) this approach (e.g. the CRediT system). True they don't have a "culture cultivation" contribution, but something like that could added pretty simply I think. I like how it's an explicit, versatile expression of values

Not AWS specifically, but There is the argument that dual use (with clear pathway to beneficial impact) is justifiable when there is relatively little beneficial use relative to harmful use, but I think its validity has to be case by case. Weird that technical success is the signal to harmful dev.

If there is anything worse than "economist brain", it's economist brain arguments about how inefficient the state is.

this is somehow is a subtweet

Too bad people didn't take Bard Ermentrout seriously when he proposed synchronizing intervention experiments in stadium bathroom systems.

Three images of street art in Pilsen. First shows Superman punching an ICE agent in paramilitary uniform. Second shows that somebody covered over Superman with black paint Third shows the artist went back and, in place of Superman, he painted Superman's dog Krypto pissing on the ICE agent.

1800 block of South Western Avenue in Chicago

Ironic how that last sentance in the image (as a belief) is a deeply held fiction for certain individuals ('deep state is after me') who voted for a govmnt that's doing just that to other people whose idealistic notion of govmnt presumably made them not believe it could happen, least of all to them.

Interesting analysis that reminds me that political discourse is a technology of control in democracy that is used to stabilize ideologies.

Does the model you are describing here that includes discursive constructions have a name/has been studied?

Or caused by strategic manoeuvring in some abstract space of ideological competition. Individuals espousing the view may well believe it as a core value, just unaware or ambivalent that it is strategic.

Partisan sorting claims another victim

I don't like where this is going...the distinction is nuance and the modern media bit just isn't equipped for nuance.

He's not playing to succeed as a career politician. He's playing another game, the rewards of which are correlated with his destructive picks/approach to governance. It's probably not productive to understand what he's doing as strategic with respect to governance either. It's another game.

A fun reminder that a loss of 60,000 jobs in and around academia is ~50% more than all coal mining jobs in the entire country.

Motherfucking wind farms…

Welcome 😁

Listings! Bring them back!! (Also a wonderful love letter to NYC at a certain formative time.)

Most of the internet used to be like this. This is actually the default, it took companies enclosing the internet and adding weird, soul-killing incentives to make people behave the way they do now. In a way, there is truly nothing special about Wikipedia except that it survived longer.

Opinion dynamics: Statistical physics and beyond Opinion dynamics, the study of how individual beliefs and collective public opinion evolve, is a fertile domain for applying statistical physics to complex social phenomena. Like physical systems, soc...

Review "Opinion dynamics: Statistical physics and beyond"
arxiv.org/abs/2507.11521

Lab experiments, data, models, analytical/computational tools. 93 pages, >1k references.

With Fabian Baumann, David Garcia, Gerardo Iñiguez, Márton Karsai, Jan Lorenz, Katarzyna Sznajd-Weron. Led by Michele Starnini

Beautiful! Reminds me of a (west coast ?) version of cap d'éternité commons.wikimedia.org/wiki/File:Ca...

If you are interested in a unified collection of common misinformation detection benchmarks, check out our recent repo @hf.co