PluginRSI: Recursive Improvement of Agent Harnesses with Reusable Plugins Paper • 2609.32423 • Published 12 days ago • 8
GraphForge: Training Working Agents with Graph-Anchored Workspace Synthesis Paper • 2609.38923 • Published 8 days ago • 147
JEPA-Anything: Learning Predictive Models across Different Worlds Paper • 2609.20800 • Published 21 days ago • 77