the greatest joy of being a computational scientist is having the computer work for you while you do something else

“Interpretability plays a special role in machine learning because instead of focusing on making the AI smarter, we focus on improving human insight. I think this is the most important category of interpretability research, and we do not do enough of it.” 😎😎😎

remember to always include a QR code on your poster. spotted at #eurips

What coding with an LLM feels like sometimes.

found a file from PhD days with the FORTY-EIGHT ways "ACE inhibitor" was encoded in the EHR system we were working wth

finally got around to booking my travel for #EurIPS2025! Looking forward to connecting with the European ML scene in Copenhagen

uv is so good

Some papers really have a good intro

The more rigorous peer review happens in conversations and reading groups after the paper is out with reputational costs for publishing bad work

I'll admit, I was skeptical when they said Gemini was just like a bunch of PhDs. But I gotta admit they nailed it.

what is the purpose of VQA datasets where text-only models do better than random?

lads can we stop

quick diagram of Bluesky’s architecture and why it’s nicer here

it's frustrating how inefficient review assignments are: we target a minimum number of completed reviews per paper but in accounting for inevitable no-shows, some people end up doing technically unnecessary (if still beneficial) reviews

New work from my team! arxiv.org/abs/2507.12950 Intersecting mechanistic interpretability and health AI 😎 We trained and interpreted sparse autoencoders on MAIRA-2, our radiology MLLM. We found a range of human-interpretable radiology reporting concepts, but also many uninterpretable SAE features.

We're excited to announce a second physical location for NeurIPS 2025, in Mexico City, which we hope will address concerns around skyrocketing attendance and difficulties in travel visas that some attendees have experienced in previous years. Read more in our blog: blog.neurips.cc/2025/07/16/n...

During the last couple of years, we have read a lot of papers on explainability and often felt that something was fundamentally missing🤔 This led us to write a position paper (accepted at #ICML2025) that attempts to identify the problem and to propose a solution. arxiv.org/abs/2402.02870 👇🧵

ExplainableAI has long frustrated me by lacking a clear theory of what an explanation should do. Improve use of a model for what? How? Given a task what's max effect explanation could have? It's complicated bc most methods are functions of features & prediction but not true state being predicted 1/

Pleased to share our ICML Spotlight with @eberleoliver.bsky.social, Thomas McGee, Hamza Giaffar, @taylorwwebb.bsky.social. Position: We Need An Algorithmic Understanding of Generative AI What algorithms do LLMs actually learn and use to solve problems?🧵1/n openreview.net/forum?id=eax...

This is cool arxiv.org/abs/2505.18235. Linear representation hypothesis discourse needs more differential geometry i m o

🚀 After two+ years of intense research, we’re thrilled to introduce Skala — a scalable deep learning density functional that hits chemical accuracy on atomization energies and matches hybrid-level accuracy on main group chemistry — all at the cost of semi-local DFT ⚛️🔥🧪🧬

Limited time offer: I get an email every time someone fills in this form, so act now to add to the chaos of my inbox before I figure out how to turn this off

A strange thing about living near Duxford is having WW2-era planes flying overhead on a regular basis.

90% of doing science is being open to new ideas.

NeurIPS 2025 call for ethics reviewers neurips.cc/Conferences/... #neurips2025

I am disappointed in the AI discourse steveklabnik.com/writing/i-am...

What happens in SAIL 2025 stays in SAIL 2025 -- except for these anonymized hot takes! 🔥 Jotted down 17 de-identified quotes on AI and medicine from medical executives, journal editors, and academics in off-the-record discussions in Puerto Rico irenechen.net/sail2025/

Registration is now open for #CHIL2025! This year's program will feature 🔹Keynote presentations 🔹Panel discussions 🔹Year in Review 🔹Posters sessions 🔹Doctoral Symposium lightning talks 👉 Schedule: chil.ahli.cc/attend/sched... 👉 Register: ahli.cc/chil25-regis...

I wrote something up for AI people who want to get into bluesky and either couldn't assemble an exciting feed or gave up doomscrolling when their Following feed switched to talking politics 24/7.

taking a break from job 2 (roller derby spreadsheets) to work on job 3 (reviewing papers)

found a copy of the NeurIPS 1989 proceedings at the office

One of my grand interpretability goals is to improve human scientific understanding by analyzing scientific discovery models, but this is the most convincing case yet that we CAN learn from model interpretation: Chess grandmasters learned new play concepts from AlphaZero's internal representations.

I genuinely think about this book all the time

Happy to announce a collaboration with the Mayo Clinic to advance our research in radiology report generation! newsnetwork.mayoclinic.org/discussion/m... Tagging some of the core team: @valesalvatelli.bsky.social @fepegar.com @maxilse.bsky.social @sambondtaylor.bsky.social @anton-sc.bsky.social

me, eating potato waffles straight from the toaster: see, vegan diets are healthier

enjoying listening to the BBC Reith Lectures (www.bbc.co.uk/programmes/b...). In the 2021 edition, Stuart Russell offhandedly predicted AlphaFold winning the Nobel prize!

Working on a "special" kind of data (health; medical images etc.) means experiencing the disappointment of realising the method in a paper relies critically on "having a very good image captioner", "asking GPT-4V", etc.

I love having a very specific question I'm trying to answer and just going to town on semantic scholar.

Let us start 2025 in a positive mood: here are 10 methods things researchers can worry *less* about in 2025

THREAD This Is The Story Of The The Pernicious Rise of AI-Generated Papers and their Online Impact An Incomplete History Told In The Voice of Documentarian Adam Curtis 1/31

feeling a but under the weather this week … thus an increased level of activity on social media and blog: kyunghyuncho.me/i-sensed-anx...

A word of caution: it might feel like that grant/paper/review cannot possibly wait over the Christmas period, but you can't get time with friends and family back. You likely won't be like "really wish I'd submitted that" in 5 years time but you might regret not spending time with those you love.

Microsoft's Computational Social Science group may have the opportunity to hire one researcher Senior: 0-3 yrs post PhD jobs.careers.microsoft.com/global/en/jo... Principal: 3+ yrs post PhD jobs.careers.microsoft.com/global/en/sh... Please note: our ability to hire this season is not certain

Isaac Kohane highlighting that AI is already being used by health insurance companies to deny claims #ML4H

Simulating a skin lesion dataset for evaluating dermatology models, from Adarsh Subbaswamy at ML4H

“Most people who have died from malaria were never in an EHR system” - Megan Coffee on the panel (“The Promise of Foundation Models in Healthcare”) at ML4H

Max Tegmark pointing out the benefits of “tool AI” over “AGI” at the #NeurIPS2024 Safe Generative AI workshop

my roller derby name is Kernel Panic 💅

It’s kicking off in the #NeurIPS2024 town hall slido

Rosalind Picard on involving the real people who will be impacted/benefit from your research *in your research*. (One of many excellent lessons from her talk on “How to optimise what matters most?”) #NeurIPS2024