DeskForge: Dense Supervision from Desktop Environments for Computer-Use Agents Paper • 2610.02320 • Published 4 days ago • 2
Moving Beyond Sparse Grounding with Complete Screen Parsing Supervision Paper • 2602.14276 • Published Feb 15 • 1
SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion Paper • 2503.11576 • Published Mar 14, 2025 • 177