ClawBench: Can AI Agents Complete Everyday Online Tasks? Paper • 2604.08523 • Published 3 days ago • 102
ClawBench: Can AI Agents Complete Everyday Online Tasks? Paper • 2604.08523 • Published 3 days ago • 102
Watch Before You Answer: Learning from Visually Grounded Post-Training Paper • 2604.05117 • Published 6 days ago • 31
Watch Before You Answer: Learning from Visually Grounded Post-Training Paper • 2604.05117 • Published 6 days ago • 31
PaperBanana: Automating Academic Illustration for AI Scientists Paper • 2601.23265 • Published Jan 30 • 223