What is the role of memory for modeling time dependent PDEs? I will be at ICLR presenting our paper (Oral) where we study when it is beneficial for modeling time-dependent PDEs! 🔗openreview.net/forum?id=o9kqa… [Oral]: Thu 24 Apr 10:30 am @ Session 1E [Poster]: Thu 24 Apr 3 pm #617

🚨 Final Reminder: Encode Fellowship Applications Close Tomorrow! (Wed Apr 30)

SymbolicRegression.jl v1.10.0 is out! It can now evolve expressions over arbitrary input types. The video below shows it reverse-engineering a string transformation from examples. Curious to see how people use this!

Wow. Apparently some PySR users are building dedicated rigs for faster equation discovery!

Has anything been written about Google's LLM turnaround? They seemed to be lagging for a while, but the recent Gemini models are super impressive. I would be interested to read an analysis of what changed in their approach or team structure.

I am a HUGE fan of Mooncake.jl github.com/chalk-lab/Moon…, a new language-level autodiff for Julia. Most impressive is how uncompromising the devs are about correctness and reliability. It has some of the best software engineering practices I’ve seen in the entire ecosystem!

Tomorrow I get to give my favourite type of lecture 😃

Really loving AR glasses for working while traveling. No squinting at my laptop + I can still enjoy the view! These just plug into my closed laptop as an external monitor. Means I'm actually productive on planes now because I'm not hurting my neck looking down the whole flight!

Closed source is so lame

There is an opening for a University Assistant Professor in Machine Learning to be based at Cambridge MLG jobs.cam.ac.uk/job/49361/ Apply!

What are the odds that LLMs which conceal their chains of thought are doing all their "thinking" in a mix of languages? Then having a second LLM translate it back to the user

Awesome workflow for building user-friendly software: - Get an LLM to call your API - Note every mistake as a potential pitfall for users - Ensure such misuses are covered by helpful error messages!

“We can’t open source our models because of risks” but surely open sourcing the algorithms involved in training these models should be good for science and safety, right? Right?

Cannot wait for someone to wrap SAP Concur in an MCP It’s kind of crazy you still can’t generate expense reports from a folder of receipts!

Apache 2.0 > BSD-3 > MIT I think there are very few reasons not to choose Apache 2.0 out of these three. Has better legal protections for developers and users (patent clause), and is just as permissive