Morning Edition · Monday, July 20, 2026Published at 1:31 AM EDT · New York

Study Finds a 'Global Workspace' of Verbalizable Representations Inside Language Models

The paper argues that only a subset of a model's internal representations are available for verbal report and flexible reasoning, borrowing a framework from cognitive neuroscience.

A new interpretability paper presents evidence that language models contain a structured subset of representations that are verbalizable, meaning available for verbal report, deliberate control, and flexible reasoning, in a way the authors…