Very interesting thread on, among other things, human verification and understanding of the latest AI proofs

Henry Yuen@henryyuen.bsky.social · 4d ago

Some initial thoughts, and a complicated mix of feelings. Wow. I mean, Erdos problems are cool (I genuinely mean that), I didn't know about the Jacobian conjecture before it got disproved. But this newest batch from OpenAI hits home in a way the previous announcements did not.

Would you like to try the first ever algorithmic feed with CONTROLS? GreenEarth is built to serve you, not the platform or anybody else. Now in beta. www.greenearth.social/p/welcome-to...

Welcome to the GreenEarth Feeds

Social media algorithms built for people who hate social media algorithms

greenearth.social

I'm delighted that Dr. Penelope Lewis is joining arXiv as our CEO!

Richard Sever@richardsever.bsky.social · 6d ago

arXiv has spun out as an independent non-profit like openRxiv. Today they announce their new CEO and founding Board blog.arxiv.org/2026/07/30/a...

Thrilled to share that I am joining UC Berkeley as an Assistant Professor in the School of Information! I start in Fall 2027, and I am recruiting PhD students this cycle. List me in your application if you want to work with me! More on what I'm looking for (and a form to indicate interest) 🧵

At my first #IC2S2 this week -- would be great to meet and chat with folks here!

Looking forward to giving a tutorial on 𝗦𝗼𝗰𝗶𝗮𝗹 𝗠𝗲𝗱𝗶𝗮 𝗙𝗲𝗲𝗱 𝗥𝗮𝗻𝗸𝗶𝗻𝗴 𝗔𝗹𝗴𝗼𝗿𝗶𝘁𝗵𝗺𝘀 with @tiziano.bsky.social at @ic2s2.bsky.social! 🕜 𝗧𝗶𝗺𝗲: Tuesday, 1-4 pm 📍 𝗟𝗼𝗰𝗮𝘁𝗶𝗼𝗻: Mansfield (210). 🌐 𝗪𝗲𝗯𝘀𝗶𝘁𝗲: social-media-ais.github.io/ic2s2-26-tut...

Do you like your social media less toxic? Yeah, we do too. @eurosky.social and Germany's Federal Agency for Breakthrough Innovation (SPRIND) just signed an agreement to collaborate on precisely that: eurosky.tech/news/eurosky...

Eurosky is collaborating with SPRIND to test a non-toxic algorithm for the first time - Eurosky

Eurosky is a European initiative to build and operate sovereign social web infrastructure

eurosky.tech

I've been a @zotero.org user for more than 6 years and have thousands of references - I've always dreamed of being able to see who else was reading those references on my virtual bookshelves. That dream is starting to come true ✨ With Zemble, every reference becomes a portal 🌀 >

𝕮@chrisshank.com · 3w ago

Just released an early preview of Zemble, a @zotero.org plugin for @semble.so! - Augment you Zotero library with all of the activity happening on Semble/Atmosphere. - Easily save and view items in Zotero on Semble. - Publish Zotero collections as Semble collections.

Pleased to support Mayor Mamdani's PIT Crew initiative. Done right, digital government saves people time and money. Done wrong, it can be costly, useless, or discriminatory. How a city interacts with its people shouldn't be outsourced. It should be local and accountable. www.nyc.gov/mayors-offic...

Mayor Mamdani Launches "Public Interest Technology (PIT) Crew" to Rapidly Build Digital Solutions to Public Problems

nyc.gov

🧵Can we reconcile excitement for SAEs with negative results? Our #ICML2026 position paper argues that even if SAEs underperform baselines when acting on knowns (e.g. probing, steering), they're a powerful tool for ~discovering unknowns~ Poster: Tue 2pm, HALL A #1715

This workshop starts at a couple of hours in case you're here at EC!

Nikhil Garg@nkgarg.bsky.social · 3mo ago

We are very excited to announce our first workshop on From Theory to Practice: behind the scenes on research deployments at EC’26 (July 6 in Rome)! Call for posters and submissions now open! Organized by myself, @ericachiang.bsky.social , Bailey Flanigan, @brwilder.bsky.social

On Monday @ncollina.bsky.social @iraglobusharris.bsky.social and I are giving a tutorial at ICML on (multi)calibration and its applications. You can find slides and an annotated bibliography on the website: calibration-tutorial.github.io as well as an interactive demo of online calibration algs.

Calibration, Decisions, and Collaboration in Learning | ICML 2026

An ICML 2026 tutorial on making probabilistic predictions trustworthy for downstream decision-making and collaboration.

calibration-tutorial.github.io

1/7 Excited to share EconCSLib: a Lean library/workflow. Our vision is to enable researchers who don't know Lean to formalize their Applied Modeling papers. Why? I, and most theorists, are terrified of fatal mistakes in our papers Paper: arxiv.org/abs/2606.13306 Project: gargnikhil.com/EconCSLib/

I am excited to be at @facct.bsky.social this year presenting a new 📝 "Scrutinizing Index-Based Risk Assessments: A Case Study in NYC Decision-making for Heat Emergency Management" (work with Luke Boyce, @angelinawang.bsky.social , and @allisonkoe.bsky.social ). 🔗: dl.acm.org/doi/10.1145/... (1/10)

🧵 How do companies in highly regulated domains actually test their AI systems for and mitigate discrimination? Our new @upturn.org paper at @facct.bsky.social w/ Emily Black, @mbogen.bsky.social, @s010n.bsky.social, and @wesleydeng.bsky.social, tries to answer this question: arxiv.org/abs/2606.02957

We don’t always know what problems are hard for LLMs. So devs evaluate on tasks HUMANS find hard or on broad benchmarks. What if we could instead anticipate which scenarios a model will fail on—all without evaluating specific input examples? 🧵NEW PAPER by @jenniferlumeng.bsky.social

Modern LLMs are incredibly good compression algorithms, which can shed light on why autonomous data science agents don't overfit as much as you might think. arxiv.org/abs/2606.11045

David Shmoys wrote an op-ed in the SF chronicle, based on our paper on how multimember districts with ranked choice voting would several limit gerrymandering www.sfchronicle.com/opinion/open...

The partisan gerrymandering war is out of control. Here’s how we can end it

OPINION: To end partisan gerrymandering, the U.S. should redraw Congressional districts, making them larger and so that each one would have multiple representatives.

sfchronicle.com

✨New paper out @nature.com ✨ For 8 weeks around the 2024 US election, we randomly assigned 2,000 people to use social media algos we built ourselves. Do engagement-based algorithms amplify intergroup, moral & emotional (IME) content—and does that distort how we see political norms? 🧵🔗 👇

This is the best thing I've read on baseball's new ABS system. Balls and strikes are made, not found. arxiv.org/pdf/2605.16237

an agent system and a single 404 error got me in trouble with OpenAI and Cornell campus security. how did this happen? introducing: "agent meltdowns" (1/11) arxiv.org/abs/2605.19149

Excited to share our paper! Due Process on Hold: A Queueing Framework for Improving Access in SNAP arxiv.org/abs/2605.15165 Millions of Americans interface with the social safety net via call centers that are too congested. In Holmes v. Knodell, bad operations = procedural due process violation.

Due Process on Hold: A Queueing Framework for Improving Access in SNAP

The U.S. social safety net delivers essential services at mass scale, but access burdens persist, as congested contact or call centers serve as a primary mode of application completion and assistance....

arxiv.org

Giving a "pub talk" tomorrow evening thanks to @pintofscience.us. (And folks talking all week, e.g. see @informor.bsky.social on Wednesday). www.pintofscience.us/event/who-de...

Who Decides? Law, Values, and Algorithms in an Automated World

From courtrooms to code, how much should we trust algorithms? Explore why judges using LLMs may be misguided, how values are socially constructed, and how …

pintofscience.us

Mor Naaman@informor.bsky.social · 3mo ago

Hey NYC friends, this is something really cool and unusual that I am doing in a couple of weeks -- a public pub talk about my research! Thanks to @pintofscience.us for organizing. Wednesday, May 20th at 7pm (it's $2 plus the beer you will be having). pintofscience.us/event/smart-...

1/ New @Nature! We study how powerful institutions shape the information environment for LLMs. Commercial LLM training is opaque, so we trace a path from state-coordinated media -> training data -> model responses.

We are very excited to announce our first workshop on From Theory to Practice: behind the scenes on research deployments at EC’26 (July 6 in Rome)! Call for posters and submissions now open! Organized by myself, @ericachiang.bsky.social , Bailey Flanigan, @brwilder.bsky.social

A post from @wtgowers.bsky.social on math research with reasoning LLMs (e.g., ChatGPT 5.5-Pro). Having similar experiences and agree with his reflections. gowers.wordpress.com/2026/05/08/a...

A recent experience with ChatGPT 5.5 Pro

We are all having to keep revising upwards our assessments of the mathematical capabilities of large language models. I have just made a fairly large revision as a result of ChatGPT 5.5 Pro, to whi…

gowers.wordpress.com

Add Bluesky discussion to academic papers: 1. enter the arxiv url here: pubchat.org 2. replace the arxiv link on your webpage with the resulting PubChat link. 3. comments added via Bluesky are aggregated on your paper's PubChat page.

PubChat

Attach Bluesky discussion to academic papers.

pubchat.org

As an avid word-game player, I think @kennypeng.bsky.social and Nicholas built a really fun one! Also some fun "small worlds" math underneath. Please play and let us know how you like it!

Kenny Peng@kennypeng.bsky.social · 3mo ago

We made traversle.io, a new daily word game! The goal is to traverse from a start word to a target word through a network of related words. (Our motivating question: is it possible to construct a network that allows human navigation?)

The EAAMO conference deadline is coming up! conference.eaamo.org/cfp/ Great community at intersection of CS-Operations-Econ and social good. Flexible publication format (e.g., non-archival option) and so costless to submit here as well!

What are the worst ethical disasters in NLP history? (I'm teaching "ethics of NLP" tomorrow and history is good for teaching this topic.) Most are data breaches/releases (AOL search logs, OKCupid profiles, Finnish therapy records...) but what others? I'll put some other examples in thread --> 1/n