GlobeBench: Evaluating Text-Based World Model Capabilities for Scalable Environment Simulations Paper โข 2610.22708 โข Published Oct 15 โข 6
FigmaTrace: Capturing Creative Nuances in Human Figma Design Workflows Paper โข 2608.21460 โข Published Aug 20
Masked Diffusion Language Models are Strong and Steerable Text-Based World Models for Agentic RL Paper โข 2607.16204 โข Published May 7 โข 7
Benchmarking Reward Hack Detection in Code Environments via Contrastive Analysis Paper โข 2601.20103 โข Published Jan 27 โข 2
GlobeBench: Evaluating Text-Based World Model Capabilities for Scalable Environment Simulations Paper โข 2610.22708 โข Published Oct 15 โข 6