Curious as to if people think if (when?) ‘superhuman AI’ arrives, will the building blocks of its performance be human recognizable concepts which have been applied and combined in new and novel ways to achieve ‘superhuman’ performance? Or will it be completely uninterpretable?

UT Austin Linguistics is hiring in computational linguistics! Asst or Assoc. We have a thriving group sites.utexas.edu/compling/ and a long proud history in the space. (For instance, fun fact, Jeff Elman was a UT Austin Linguistics Ph.D.) faculty.utexas.edu/career/170793 🤘

I will be giving a short talk on this work at the COLM Interplay workshop on Friday (also to appear at EMNLP)! Will be in Montreal all week and excited to chat about LM interpretability + its interaction with human cognition and ling theory.

The compling group at UT Austin (sites.utexas.edu/compling/) is looking for PhD students! Come join me, @kmahowald.bsky.social, and @jessyjli.bsky.social as we tackle interesting research questions at the intersection of ling, cogsci, and ai! Some topics I am particularly interested in:

Breaking my years-long vow to never fly American Airlines just to be met with a 6 hr delay and 5am arrival back home 🫠

A key hypothesis in the history of linguistics is that different constructions share underlying structure. We take advantage of recent advances in mechanistic interpretability to test this hypothesis in Language Models. New work with @kmahowald.bsky.social and @cgpotts.bsky.social! 🧵👇!

LMs learn argument-based preferences for dative constructions (preferring recipient first when it’s shorter), consistent with humans. Is this from memorizing preferences in training? New paper w/ @kanishka.bsky.social , @weissweiler.bsky.social , @kmahowald.bsky.social arxiv.org/abs/2503.20850

New preprint w/ @jennhu.bsky.social @kmahowald.bsky.social : Can LLMs introspect about their knowledge of language? Across models and domains, we did not find evidence that LLMs have privileged access to their own predictions. 🧵(1/8)

Notoriously finicky BC weather celebrating the last day of #NeurIPS2024 with a rainbow across the harbor

I'll be presenting a position piece (arxiv.org/abs/2409.17005) on what cognitive science and linguistics can bring to the Math + AI field tomorrow from 11:00 -12:30 and 4:00 - 5:00 at the (aptly named) #NeurIPS2024 MathAI Workshop in West Meeting Room 118-120. Come say hi and hear about my work!

In Vancouver for #NeurIPS2024 workshops! At Math-AI tomorrow @sashaboguraev.bsky.social is presenting our experiment-infused position piece on the communicative nature of math and why that matters for AI arxiv.org/pdf/2409.17005. Say hi! Will be better than the Panthers 4-0 loss to the Canucks.

I’m at NeurIPS all week! On 12/14 I’ll present on viewing math as a communicative activity at the MathAI workshop. Meanwhile, I’d love to chat about this work or more broadly about communicative framing in NLP, including emergent communication paradigms, language games and more. Please reach out!