James Michaelov
77 posts
@jamichaelov
Postdoc @MIT. Previously: @CogSciUCSD, @CARTAUCSD, @AmazonScience, @InfAtEd, @SchoolofPPLS. Research: Language in humans and machines
Excited to announce that I’ll be presenting a paper at #NeurIPS this year! Reach out if you’re interested in chatting about LM training dynamics, architectural differences, shortcuts/heuristics, or anything at the CogSci/NLP/AI interface in general! #Neurips2025
Excited to announce that I’ll be presenting a paper at #NeurIPS this year! Reach out if you’re interested in chatting about LM training dynamics, architectural differences, shortcuts/heuristics, or anything at the CogSci/NLP/AI interface in general! #Neurips2025
@jamichaelov and I will be presenting our paper at the @CogInterp workshop 13:15 - 14:45 on Dec 7th. We show how disaggregating grammatical benchmarks over the course of training reveals stages of training where models learn heuristics before learning more generalizable patterns.
✨New pre-print✨ Crosslingual transfer allows models to leverage their representations for one language to improve performance on another language. We characterize the acquisition of shared representations in order to better understand how and when crosslingual transfer happens.
Excited to present this at COLM this week! Reach out if you want to meet/chat!
New preprint with @linguist_cat and Ben Bergen! We’ve all heard of the new wave of recurrent language models, but how good are they for modeling human language comprehension? Quite good, it turns out! 🧵 arxiv.org/abs/2404.19178
New preprint with @linguist_cat and Ben Bergen! We’ve all heard of the new wave of recurrent language models, but how good are they for modeling human language comprehension? Quite good, it turns out! 🧵 arxiv.org/abs/2404.19178
5️⃣Michaelov etal. find surprisal explains N400s to sentence-final words varying in predictability, plausibility, and relation to the likely completion better than sem. similarity. The results support lexical predictive coding accounts. shorturl.at/knsRY @jamichaelov 7/n
NAACL 2024 seems to charge $750 for students to register if they're a presenter (every paper requires at least one registered presenter). @naacl am I reading this right? Seems like a major burden on students, especially if (as is common) only a paper's student authors attend.
👶 BabyLM Challenge is back! Can you improve pretraining with a small data budget? BabyLMs for better LLMs & for understanding how humans learn from 100M words New: How vision affects learning Bring your own data Paper track babylm.github.io 🧵
Just realized I haven’t really posted much about my research in the last few years, so I think I'm going to do a bit of a catch-up
