Sebastian Schuster

Following long contexts in natural language requires tracking what happens to entities that have been mentioned. Can LLMs do this as new information about them is presented in text? In @najoungkim’s and my #ACL2023 paper, we present a behavioral task that targets this question.

RExBench is now available in Terminal Bench (@harborframework)! 🎉 We integrate 2 tasks (cogs, othello) along with a local testing framework so you can test if your agents can autonomously implement novel AI research extensions.

@najoungkim do you?

i think about this spam title from time to time

🧵 Do coding agents know when to ask for help? Real-world coding tasks are rarely fully specified, yet most agents are optimized to execute autonomously rather than clarify.

I think we're going to need CS PhD students to do far more than provide accountability, by which I think Sayash means do code review for AI agents and make sure the agent isn't making silly mistakes. The main value of a strong PhD student for a PI is that they're immersed in a problem, a method, an application, a collaboration with another field; they are obsessed with finding the next question to ask, not just executing the experiments their advisor asks them to do. I simply wouldn't be able to work on the range of things I'm able to work on if I were going it on my own, even if all of my code was generated instantaneously by an agent.

Diffusion LLMs can think EoS-by-EoS! The higher the generation length, the better the performance of Masked Diffusion LLMs, even though they generate the same amount of words and only augment them with more and more EoS tokens  👀

📢 Life update 📢 After a wonderful time at @allen_ai, I've joined @CisLmu at @LMU_Muenchen as a tenure-track assistant professor in NLP. Thrilled to be back in Europe and to start a lab in Munich's flourishing AI ecosystem! 🎉

Go work with Najoung!

🎓🤖🧠 Fully funded 4-year PhD in Computational Linguistics @ucl - @UCLBrainScience - Division of Psychology & Language Sciences New lab launching. Open topic. Strong computational focus. 🗓 Apply by 7 January 🇬🇧 UK home-rate students only (unfortunately) Details ⬇️

🧑‍🔬I’m recruiting PhD students in Natural Language Processing @UniLeipzig Computer Science, together with @Sca_DS! Topics include, but aren’t limited to: 🔎Linguistic Interpretability 🌍Multilingual Evaluation 📖Computational Typology Please share! #NLProc #NLP

At the #Neurips2025 mechanistic interpretability workshop I gave a brief talk about Venetian glassmaking, since I think we face a similar moment in AI research today. Here is a blog post summarizing the talk: davidbau.com/archives/2025/…

@aryaman2020 ICLR and NeurIPS reviewers this year really make ACL reviewers shine

i hate ML conference reviewers. i take back everything bad i ever said about ACL. every ACL reviewer i ever got was at least literate

Personally, @chrmanning is such an inspiration and someone I have unfalteringly admired in the past 15 years or so of working in NLP. Imagine producing phd students who have, in their own right, become stars, repeatedly producing test-of-time science, being responsible for companies that are collectively worth billions, and yet deeply technical and modest in in-person conversations. Impossible not admire and emulate.

If you want to learn AI from the experts, keep reading. 💡 Together with @UCL, we made a free AI Research Foundations curriculum – available now on Google Skills. With lessons from a Gemini Lead like @OriolVinyalsML, you'll explore how to code better, fine-tune an AI model and more. → skills.google/collections/de…

I’m recruiting PhD students for 2026! If you are interested in robustness, training dynamics, interpretability for scientific understanding, or the science of LLM analysis you should apply. BU is building a huge LLM analysis/interp group and you’ll be joining at the ground floor.

I am hiring a PhD student to start my lab at @ucl! Get in touch if you have any questions, the deadline to apply through ELLIS is 31 October. More details🧵

@najoungkim Missed opportunity to use an em-dash in this tweet

humans seem to have a strong urge to make themselves distinct from ai even actively changing how they talk/write, maybe this'd be why they'd never become indistinguishable (tho if ai picks up on this and tries to distance from self that'd be interesting but there'd still be a lag

@mtoneva1 So cool, congratulations!!

So excited and honored to receive an ERC Starting Grant for the project BrainAlign!! BrainAlign will bring LLMs closer to human understanding by directly aligning them with the human brain. Stay tuned for our findings, and multiple postdoc and PhD openings in the coming years!

faculty opening in computational linguistics and cognitive models of language and cognition at the University of Trento (deadline Sep 14): cimec.unitn.it/en/71/language… I visited this department a couple of times and loved the science, the city, and the proximity to lago di garda!

Work with me and @najoungkim on a cool project evaluating the latest agents. Joining us at BU would be preferable but if you make a strong case, we **may** also be able to hire you in Vienna. #nlproc

I'm grateful for the recognition, it's was great to work on this project with @michahu8, @jowenpetty, Chuan Shi and @lambdaviking!