Mage-VL-Cowork: A Full Training Recipe to Learn Hybrid Computer Use From Evolved Harnesses Paper • 2610.40747 • Published 23 days ago • 24
ParetoSaddler: Pareto-Guided Optimization of Agent Harnesses for Quality and Cost Paper • 2611.02079 • Published 20 days ago • 22
WebHarbor: Docking Real Websites for Evolving GUI Agent Environments Paper • 2611.03634 • Published 19 days ago • 28
OpenWebRL: Demystifying Online Multi-turn Reinforcement Learning for Visual Web Agents Paper • 2606.02031 • Published Jun 1 • 21
OpenForgeRL: Train Harness-native Agents in Any Environment Paper • 2607.21557 • Published Jul 23 • 12
AutoEnvScaling: Automating the Data Flywheel with Terminal Agents Paper • 2610.14163 • Published Oct 9 • 14
Self-Evolving Coding Agents via Harness-Policy Optimization Paper • 2610.22791 • Published Oct 16 • 50
Agent World Model in the Wild: Computer-Use Agent Environment Synthesis Paper • 2610.26828 • Published Oct 19 • 72
PrismaCoder: Pushing the Boundaries of Open Multimodal Code Intelligence through Decomposed Rubric Synthesis Paper • 2610.20538 • Published Oct 14 • 57
Agent World Model in the Wild: Computer-Use Agent Environment Synthesis Paper • 2610.26828 • Published Oct 19 • 72
Self-Evolving Coding Agents via Harness-Policy Optimization Paper • 2610.22791 • Published Oct 16 • 50