Excited to share our work on the cultural adaptability of LLMs!
📣We release NormAd, a dataset to measure LLM adaptability to varying degrees of social and cultural norms from 75 different countries!
📜: arxiv.org/abs/2404.12464

#Speech people, unfortunately someone is impersonating me as ISCA president again and is asking for help/ money for #isca .

Please ignore and know that I would never do that without any explanation.

#scam 😔

On May 2-3, we're going to have a big event in Pittsburgh about LLM Agents. We have invited talks from great speakers inside and outside CMU, student research presentations and posters, tutorials and discussions! Come join us at CMU campus, and register at cmu-agent-workshop.github.io

We're having a big event on agents at CMU on May 2-3 (one week from now), all are welcome! cmu-agent-workshop.github.io

It will feature:
* Invited talks from Alane Suhr / suhr @ sigmoid . social Yu Su Xinyun Chen Maarten Sap (he/him) and Chris Paxton
* Posters of cutting edge research
* Seminars and hackathons

The latest talk at ZurichAI also exists as a video presentation since this morning. 🌂

I welcome feedback on the format, what works, and what doesn't. 🙏

youtu.be/yeEZpf4BlDA

This work was a collaboration between Saumya Gandhi, Ritu Gala (me), Vijay V., Sherry Tongshuang Wu, and Graham Neubig. You can read our paper at arxiv.org/abs/2404.14361. We've also integrated these ideas into the newest release of Prompt2Model (github.com/neulab/prompt2…)!

NEW - AI is now writing up police reports, via multi-billion dollar cop contractor and Taser maker Axon.

What could go wrong?

w/ Richard Nieva

forbes.com/sites/thomasbr…

**people* are using AI to write police documents*

and this will inevitably be disastrous for so many reasons that I'm too tired to explain

Announcing NeurIPS Preschool Track

This year, we invite preschoolers to submit machine learning research papers.

🫠

Good write-up of what we folks in the multilingualism #NLProc community have been working to avoid for years now!

when I came up with the idea of the NeurIPS high school track I didn't expect it to be met with such universal acclaim. humbling!

hear me out: OceaniACL, EurasiACL, and EastasiACL

Ahia et al. (2023; aclanthology.org/2023.emnlp-mai…) observed that the same is true for current LLMs such as ChatGPT: They segment text in non-English languages into many more tokens and are thus much more costly to use in such languages.

They call this “double unfairness”: higher prices…

The fact that it's so easy to confuse etymology and entomology is just one of those linguistic bugs

Super proud of Illinois Theorem Prover undergraduate researcher Max Fan for getting the NSF GRFP this year! Congrats! Max is currently deciding where to go for his PhD. Whoever gets to work with him is lucky 😄. PL/FM/SE at Illinois Illinois Computer Science

I am bizarrely proud of this