The Ignition Index: Measuring Global Workspace Dynamics in Language Models

arXiv:2604.28031v1 Announce Type: cross Abstract: When researchers iteratively refine ideas with large language models, do the models preserve fidelity to the original objective?

arXiv:2609.01924v1 Announce Type: new Abstract: Recent work identifies a mid-depth band of verbalisable, causally potent representations in a standard feedforward transformer ---

arXiv:2603.29735v1 Announce Type: new Abstract: The evolution of intelligence in artificial systems provides a unique opportunity to identify universal computational principles.

arXiv:2604.13206v1 Announce Type: new Abstract: As Large Language Models (LLMs) are increasingly integrated into agentic workflows, their unpredictability stemming from numerical

arXiv:2603.18532v1 Announce Type: cross Abstract: The strong performance of large vision-language models (VLMs) trained with reinforcement learning (RL) has motivated similar

<!-- SC_OFF --><div class="md"><p>&#x200b;</p> <p>## Abstract</p> <p>Post-training makes language models more decisive without necessarily making them more accurate — and we find