ICL emerges across language, genomes, proteins, and more — the Convergent Emergence Hypothesis

DanielKhashabi · x · 2026-09-17

A new paper tests whether in-context learning is unique to LLMs by training models on six very different data types — language, genomes, proteins, integer sequences, time series, and images — on the same abstract tasks:

Work led by Nate Schlueter.

Related event: JHU study finds in-context learning emerges across six modalities(4 posts)→

Original post →

More from Research

Research channel →