I’ll be at ICLR next week! 🇦🇹

Reach out if you want to talk about state-space models vs transformers, training stability, Kolmogorov-Arnold theorem, multi-modal understanding, etc.

I'll be at ICLR ICLR 2024 in Vienna next week! Let me know if you're interested in meeting up.

I'll be presenting my work on Stack Attention (spotlight paper w/ David Chiang ) on Tuesday at Poster Session 2, 4:30-6:30pm, Halle B.

Arriving in Vienna soon for #ICLR2024 ! 🙌🏼 🇦🇹

I’m so excited for my first non *CL conference, so reach out if you’d like to talk about interpretability, Humans x LLMs or just generally about music or coffee ☕️

Alsoooo, Sahana Ramnath and I will be talking about our work too ⬇️

If you're going to be in Vienna for #ICLR2024 , check out Mentoring Chats. It's a unique opportunity to chat with senior researchers and ask them about their journey and research. We have an exciting line up of mentors! See here for more details: blog.iclr.cc/2024/05/01/hug…

A good evidence for why RNN alternative transformers don't work is because google published the griffin paper. 🫣

FineWeb 🍷 is basically 𝚝𝚑𝚎_𝚒𝚗𝚝𝚎𝚛𝚗𝚎𝚝.𝚌𝚜𝚟

This is the first paper from my lab at EPFL🙂 🎊. It conducts a detailed analysis of the representation dynamics of PPO to understand the reasons for poor performance that happens under certain conditions. We also conducted detailed analyses on ways to mitigate it.

A packed room at Yoav Artzi 's keynote talk

Sasha Rush I think one nice thing is that we as a community spent a lot of time in 2023 making hard but reasonable agent benchmarks where if we make the performance go up it actually might mean something, and now we get to improve them!

Highly recommended ⬇️

Our co-founder and CEO, Demi Guo, was just named 'One to Watch' by Bloomberg Bloomberg Technology, and we agree. Keep an eye on us for what’s next 👀

Slowly but surely making progress on Conference on Language Modeling reviewing. Reviews are due May 10! 😱🤗
9%|🦙▐▐▐▐▐▐▐▐▐▐▐▐▐▐▐▐▐▐▐|

[no idea what's going on with these llamas frantically working, but I am scared]

Not really sure what it's for, but it's very fun.

CIFAR researcher Christopher Manning has been awarded the 2024 IEEE Awards John von Neumann Medal for outstanding achievements in computer-related science and technology.

Learn more about his career and life-long love of language: engineering.stanford.edu/magazine/layin…

🇦🇹 I’m going to #iclr2024 in Vienna next week. Who all do I know that’ll be there?

Students’ papers there:
Katherine Tian: arxiv.org/abs/2311.08401
Charlotte Nicks: openreview.net/forum?id=4eJDM…
Eric: arxiv.org/abs/2310.12962
Parth Sarthi: arxiv.org/abs/2401.18059

Honored to have been named the inaugural Andrew H. and Ann R. Tish Assistant Professor at Cornell Tech!

AmpleGCG: Learning a Universal and Transferable Generative Model of Adversarial Suffixes for Jailbreaking Both Open and Closed LLMs

'we first discuss the drawbacks of solely picking the suffix with the lowest loss during GCG optimization for jailbreaking and uncover the missed…

It was a mistake to pick a fight with the prompters. My replies are filled with dozens of slight variants of the same comeback. Think they're getting close.