James Michaelov (@jamichaelov) 's Twitter Profile
James Michaelov

@jamichaelov

Postdoc @MIT. Previously: @CogSciUCSD, @CARTAUCSD, @AmazonScience, @InfAtEd, @SchoolofPPLS. Research: language comprehension, the brain, artificial intelligence

ID: 907248697062428672

linkhttps://jmichaelov.com/ calendar_today11-09-2017 14:25:20

69 Tweet

353 Followers

729 Following

James Michaelov (@jamichaelov) 's Twitter Profile Photo

The key takeaway of this study is that compared to the human cloze baseline, language models over-predict words that are either related to the most predictable next word (the 'best completion') or to the event under discussion

The key takeaway of this study is that compared to the human cloze baseline, language models over-predict words that are either related to the most predictable next word (the 'best completion') or to the event under discussion
James Michaelov (@jamichaelov) 's Twitter Profile Photo

This is concerning, and I wouldn't be surprised if it leads to some students having to withdraw their papers from the conference

James Michaelov (@jamichaelov) 's Twitter Profile Photo

This paper is now accepted to be presented at Conference on Language Modeling! Updated version is on arXiv. Feeling excited for the conference, let me know if you want to meet!

James Michaelov (@jamichaelov) 's Twitter Profile Photo

Also generally interested in chatting about cognitive modeling, scaling, and language comprehension/understanding in humans and machines! Conference on Language Modeling #COLM2024