Today in Quanta Magazine, Ben Brubaker gives a new overview of our work (w/ Ashish Sabharwal) on the expressive power of transformers with/without CoT
quantamagazine.org/how-chain-of-t…

If you want a clearly written survey on the relations between transformers and formal languages, you can't do better than this, by Lena Strobl, Will Merrill, Gail Weiss, David Chiang, and Dana Angluin: arxiv.org/abs/2311.00208

Hey! Conference on Language Modeling is recruiting reviewers!
███████░░░ 70% recruited 🚀🚀🚀
Did you get an invite? Please respond NOW!
Didn't get an invite and want to help? ❤️ please fill this form: forms.gle/8HNbdLFwhP9XLH…

I like TMLR and can't think of a reason why ACLRollingReview and Findings of the ACL couldn't be merged into something like TMLR.

Our work on 'Stack Attention' was just accepted as a ✨spotlight✨ paper at #ICLR 2024! See you in Vienna!

I’m at the 2024 LSA Annual Meeting in NYC!

On Saturday 10:30am-12pm, I’ll be presenting a poster about non-discourse-configurationality of the Kichwa language, an indigenous language of Ecuador.

Stop by our poster and let’s have a chat!

Stephen Bothwell will present his #EMNLP2023 paper on parallelism detection at 9am in the East Foyer! virtual2023.emnlp.org/paper_4901.html arxiv.org/abs/2312.00100

#EMNLP2023 Honored to present our poster at WiNLP today! Looking forward to meeting friends and chatting about HCI + NLP research
Matthias Orlikowski and I are also organizing an informal lunch for researchers working on subjectivity and disagreement
Stay tuned for more details🍽️!

Our NeurIPS paper shows transformers can be expressed in first-order logic with majority quantifiers
=> Extends the theory of the expressiveness and limitations of transformers
=> Provides a 'programming language' capturing transformers' computation
x.com/lambdaviking/s…

Took a look today and this is very interesting stuff! The lower-bound direction is particularly cool: showing how LTL and counter-free automata can be simulated in a transformer through B-RASP.