Verbalizable Representations Form a Global Workspace in Language Models

LLMs Corrupt Your Documents When You Delegate

Bielik-Q2-Sharp: A Comparative Study of Extreme 2-bit Quantization Methods for a Polish 11B Language Model

Towards Provably Unbiased LLM Judges via Bias-Bounded Evaluation

Won't Get Fooled Again: Answering Questions with False Premises

International AI Safety Report 2026

From Scale to Speed: Adaptive Test-Time Scaling for Image Editing

Tree of Thoughts: Deliberate Problem Solving with Large Language Models

Evaluating AGENTS.md: Are Repository-Level Context Files Helpful for Coding Agents?

The Prompt Report: A Systematic Survey of Prompt Engineering Techniques