VideoSeeker: Incentivizing Instance-level Video Understanding via Native Agentic Tool Invocation Paper • 2605.16079 • Published 20 days ago • 28
MolmoAct2: Action Reasoning Models for Real-world Deployment Paper • 2605.02881 • Published May 4 • 348
ClawGUI: A Unified Framework for Training, Evaluating, and Deploying GUI Agents Paper • 2604.11784 • Published Apr 13 • 143