iiinova — where your digital self keeps evolving

Reposted by Tim van Erven

Huck Bennett @huckbennett.bsky.social · 22/09/2026

Wow does the post title bury the lede! I particularly like the pictured comment.

35715

Reposted by Tim van Erven

Andrew Gordon Wilson @andrewgwils.bsky.social · 21/09/2026

I'm a Program Chair for ICML 2027. We'd love any creative suggestions for a great conference! How to manage the explosion of AI slop and insane submission growth? How to manage reviewing? How to incentivize high quality creative work? How to reduce bureaucracy and overhead?

10257

Tim van Erven @timvanerven.nl · 09/09/2026

Can Transformers be decompiled into readable programs? 📣 Tomorrow (Thursday): Michael Hahn will present at the Theory of Interpretable AI Seminar! 📅 Thursday, September 10 🕓 16:00 CET / 10:00 EST 🌐 Online tverven.github.io/tiai-seminar/ Hope to see you there!

010

Reposted by Tim van Erven

uai2026 @auai.org · 17/08/2026

Day 1 at the beautiful KIT Royal Tropical Institute.

0153

Tim van Erven @timvanerven.nl · 16/07/2026

Join us today for the Theory of Interpretable AI Seminar! 🔍 Shauli Ravfogel will present “Geometric Factual Recall in Transformers”, exploring how language models may store and retrieve factual associations through geometric structure. tverven.github.io/tiai-seminar/

031

Reposted by Tim van Erven

Eugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 13/07/2026

These are very odd failures! arxiv.org/abs/2211.00241

31016

Tim van Erven @timvanerven.nl · 18/06/2026

Just learned about this website, with open mathematical problems in machine-learning-adjacent areas such as statistics, learning theory, and optimization. E.g. many open problems from COLT are there. Cool stuff. solveall.org

solveall.org

052

Tim van Erven @timvanerven.nl · 16/06/2026

Join us this Thursday for the Theory of Interpretable AI Seminar! Donya Saless will present her COLT 2025 work on using small subsets of training data to provide robust, compact explanations for ML predictions. 16:00 CET | 10:00 EST tverven.github.io/tiai-seminar/

010

Tim van Erven @timvanerven.nl · 22/05/2026

The Three Laws of Reviewing: 1. A reviewer may not decrease the conference quality or, through inaction, allow the conference quality to come to harm. 2. A reviewer must obey the orders given them by the program chairs except where such orders would conflict with the First Law. 3. ...

120

Tim van Erven @timvanerven.nl · 19/05/2026

We have Nir Shavit speaking at the Theory of Interpretable AI seminar this Thursday! 🌟 Nir is an MIT professor and Gödel + Dijkstra Prize winner for work in distributed computing. He uses combinatorial ideas to explain neural computation via feature channels. tverven.github.io/tiai-seminar/

040

Reposted by Tim van Erven

Clément Canonne @ccanonne.github.io · 11/05/2026

List of accepted papers at #COLT2026, the Annual Conference on Learning Theory: learningtheory.org/colt2026/acc... (h/t @gautamkamath.com)

learningtheory.org

COLT 2026 - Accepted Papers

0115

Reposted by Tim van Erven

Markus Eichhorn @markuseichhorn.bsky.social · 07/05/2026

Bingo! Do I win a prize now?

Comic strip titles self-sabotaging academic habits that feel normal, featuring nine panels.

1122583

Tim van Erven @timvanerven.nl · 23/04/2026

Indeed, a must-read! An existential debate about mathematics is coming up, and now is the time to articulate the many values of mathematics. If mathematicians are seen as machines that turn coffee into theorems, we will be in a lot of trouble.

151

Reposted by Tim van Erven

Russ Poldrack @russpoldrack.org · 20/04/2026

"AI amplifies whatever is already there. Good discipline becomes great output. No discipline becomes technical debt at machine speed. Anthropic chose a direction. Go faster. Have Claude check Claude. And when it breaks, go faster still." substack.com/home/post/p-...

substack.com

Claude Code's Source: 3,167-Line Function, Regex Sentiment

Anthropic claimed 100% of Claude Code is AI-written. A source leak exposed a 3,167-line function, regex sentiment analysis, and 250K wasted API calls daily

12611

Reposted by Tim van Erven

Advait Bhat @advaitmb.bsky.social · 16/04/2026

How are AI tools changing our writing process? In our #chi2026 paper, we find a shift in the way writers write and engage with ideas: from generating ideas to reacting to the ones AI surfaces. We call it reactive writing.

174

Tim van Erven @timvanerven.nl · 07/04/2026

Looking forward to Anton Xue's talk at the Theory of Interpretable AI Seminar! 🚀 He’ll present a framework with stability guarantees for certifying feature attribution explanations and LLM reasoning chains through robust feature selection. Join us! 🧠✨ tverven.github.io/tiai-seminar/

020

Reposted by Tim van Erven

Rémi Emonet @remiemonet.bsky.social · 26/03/2026

Don't forget this!

042

Reposted by Tim van Erven

NeurIPS Conference @neuripsconf.bsky.social · 27/03/2026

We want to speak directly to the concern many of you have expressed, and we owe you a clear explanation of what happened, why it happened, and where we stand now. We understand this situation caused genuine alarm and we take that seriously.

32712

Reposted by Tim van Erven

Victor Escorcia @escorciav.bsky.social · 25/03/2026

Did I miss the discussion about this over here?

772

Reposted by Tim van Erven

NeurIPS Conference @neuripsconf.bsky.social · 23/03/2026

Following the success of the EurIPS and NeurIPS-Mexico City pilots in 2025, we are thrilled to announce two official NeurIPS 2026 satellite events for this year! These will be held in Paris, France and Atlanta, USA, respectively, running alongside the main venue in Sydney, Australia.

neurips.cc

2025 Conference

The Thirty-Ninth Annual Conference on Neural Information Processing Systems

27017

Tim van Erven @timvanerven.nl · 16/03/2026

Good advice from the university library. Applying this to all my social media as well.

021

Tim van Erven @timvanerven.nl · 04/03/2026

Upcoming talk in the Theory of Interpretable AI seminar: Federico Adolfi frames interpretability in terms of computational complexity theory.

091

Reposted by Tim van Erven

Zach Weinersmith @zachweinersmith.bsky.social · 03/03/2026

Wow. Donald Knuth on Ai for math: www-cs-faculty.stanford.edu/~knuth/paper...

www-cs-faculty.stanford.edu

69322

Tim van Erven @timvanerven.nl · 12/02/2026

Very inspiring talk today by Damien in the Theory of Interpretable AI seminar! Recording is up: youtu.be/qI4oVWhvz-4 Upcoming talks on the seminar website: tverven.github.io/tiai-seminar/

youtu.be

Theory of Interpretable AI Seminar: Damien Garreau

YouTube video by Theory of Interpretable AI Seminar

040

Tim van Erven @timvanerven.nl · 09/02/2026

Excited for Damien’s seminar talk this Thursday!🚀 Steering is an exciting area in interpretability—but how strongly should we steer? Damien will present a theory of steering strength: choosing the right magnitude of representation change—not too weak, not too strong tverven.github.io/tiai-seminar/

021

Reposted by Tim van Erven

Clément Canonne @ccanonne.github.io · 24/01/2026

Apparently the NeurIPS chairs decided to reopen the camera-ready submission to ensure "that the proceedings reflect the highest scientific quality.” @neuripsconf.bsky.social, is that true? So the solution to academic misconduct is to give the authors a chance to cover their tracks?

44812

Tim van Erven @timvanerven.nl · 15/01/2026

Video recording of Chhavi's talk in the Theory of Interpretable AI seminar is now available: youtu.be/ylW3H--g7SU Check out the seminar website for upcoming speakers: tverven.github.io/tiai-seminar/

youtu.be

Theory of Interpretable AI Seminar: Chhavi Yadav

YouTube video by Theory of Interpretable AI Seminar

020

Tim van Erven @timvanerven.nl · 09/01/2026

Looking forward to the next “Theory of Interpretable AI” seminar on January 15, where Chhavi Yadav will present "ExpProof"! A fresh take on trustworthy explanations for confidential ML models using Zero-Knowledge Proofs. Feel free to join! #interpretability #Crypto tverven.github.io/tiai-seminar/

020

Reposted by Tim van Erven

ICML Conference @icmlconf.bsky.social · 08/01/2026

New blog post (on a shiny new ICML blog!): What's New in #ICML2026 Peer Review Some highlights: - Policies to combat thinly sliced contributions - Cascading desk rejections for peer-review abuse - Reviewer reciprocity - New ways to support authors and reviewers Post: blog.icml.cc/2026/01/08/w...

1238

Tim van Erven @timvanerven.nl · 09/12/2025

Ayman will present at the Theory of Interpretability Seminar! He’ll share a new approach for optimal, interpretable decision trees with strong efficiency guarantees. 🌳✨ 🗓️ Join us to learn how AO* search leads to provably better sparse trees! tverven.github.io/tiai-seminar/

030

Tim van Erven @timvanerven.nl · 21/11/2025

When reading a large literature, it is really helpful to have opinionated opinions that help to categorize papers. One of mine for explainable AI is that methods need to address a fundamental limit in how much information can be communicated. Blog post: www.timvanerven.nl/blog/xai-com... (no math)

timvanerven.nl

The Central Challenge in Explainable AI: Channel Capacity

Explainable AI is about communication: we want to tell people how or whya machine learning model is making certain decisions. Why is this sodifficult? In this post I take an information-theoretic pers...

084

Reposted by Tim van Erven

David Mimno @dmimno.bsky.social · 14/11/2025

Our libraries are cutting staff so that Elsevier can have its 32% profit margin

110936

Reposted by Tim van Erven

Gunnar König @gunnark.bsky.social · 12/11/2025

The schedule for our Workshop on the Theory of XAI is now online! 🕰️ Dec 2, starting 9am 📍 Bella Center Copenhagen (co-located with EurIPS) 🔗 sites.google.com/view/theory-...

sites.google.com

Theory of XAI Workshop

Explainable AI (XAI) is now deployed across a wide range of settings, including high-stakes domains in which misleading explanations can cause real harm. For example, explanations are required by law ...

032

Tim van Erven @timvanerven.nl · 12/11/2025

Great seminar talk by @ulrikeluxburg.bsky.social yesterday. Here's the video if you missed it: youtu.be/zR_GvDF65OM Seminar info: tverven.github.io/tiai-seminar/

youtu.be

Theory of Interpretable AI Seminar: Ulrike von Luxburg

YouTube video by Theory of Interpretable AI Seminar

071

Tim van Erven @timvanerven.nl · 10/11/2025

Coming up tomorrow (Tuesday 11 Nov) in the Theory of Interpretability seminar: Ulrike von Luxburg will discuss why informative explanations only exist for simple functions 👀 tverven.github.io/tiai-seminar/

060

Reposted by Tim van Erven

Ulrike Luxburg @ulrikeluxburg.bsky.social · 07/11/2025

Here is a formal impossibility result for XAI: Informative Post-Hoc Explanations Only Exist for Simple Functions. I'll give an online presentation about this work next tuesday in @timvanerven.nl 's Theory of Interpretable AI Seminar: arxiv.org/abs/2508.11441 tverven.github.io/tiai-seminar/

0165

Tim van Erven @timvanerven.nl · 23/10/2025

I am recruiting via Ellis for a PhD position to develop mathematical foundations for explainable AI. Benefits: - Do new fundamental math / machine learning theory on a societally important topic - Part of a team - Good conditions in Amsterdam, NL Applications via Ellis PhD program before Oct. 31.

werkenbij.uva.nl

Vacancy — PhD Position on Mathematical Foundations for Explainable AI

Are you highly motivated to do PhD research in mathematical machine learning, with special emphasis on mathematical foundations for explainable AI? If yes, the Korteweg-de Vries Institute for Mathemat...

131

Tim van Erven @timvanerven.nl · 21/10/2025

The video for Shahaf's talk is now available: youtu.be/AsJB4kym4Tc See the seminar website for upcoming talks: tverven.github.io/tiai-seminar/

youtu.be

Theory of Interpretable AI Seminar: Shahaf Bassan

YouTube video by Theory of Interpretable AI Seminar

011

Tim van Erven @timvanerven.nl · 20/10/2025

Happening tomorrow, October 21!

020

Tim van Erven @timvanerven.nl · 16/10/2025

The next talk in the Theory of Interpretable AI seminar will be by Shahaf Bassan on "The Computational Complexity of Explaining ML Models". Tuesday October 21 tverven.github.io/tiai-seminar/

020

Reposted by Tim van Erven

ELLIS @ellis.eu · 13/10/2025

How can we make AI explanations provably correct — not just convincing? 🤔 Join us for the Theory of Explainable Machine Learning Workshop, part of the ELLIS UnConference Copenhagen 🇩🇰 on Dec 2, co-located with #EurIPS. 🕒 Call for contributions open until Oct 15 (AoE) 🔗 eurips.cc/ellis

1134