news.yale.edu web signal

How do AI chatbots ‘reason’ like humans? A Yale-led study offers clues

By Alexis Dufresne Published October 10, 2026 at 03:12 UTC

Summary

Tal Linzen @tallinzen.bsky.social amplified

@nyudatascience.bsky.social

LLMs run on vectors, not symbols. A preprint led by Yale’s Tom McCoy (@rtommccoy.bsky.social) with CDS Assoc. Prof. @tallinzen.bsky.social found those vectors closely approximate symbolic structure in language, arithmet…
View on Bluesky →
Tom McCoy @rtommccoy.bsky.social: A very nice writeup from Mike Cummings at Yale News on our work finding symbolic structure inside LLM representations! Read it if you want … →

Originally reported by news.yale.edu

Original headline: How do AI chatbots ‘reason’ like humans? A Yale-led study offers clues