GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed Paper • 2609.39600 • Published 25 days ago • 58
GraphForge: Training Working Agents with Graph-Anchored Workspace Synthesis Paper • 2609.38923 • Published 25 days ago • 78
vmachadocos/visual-question-answering-study31 Visual Question Answering • 16.6k • Updated Sep 4 • 7
vmachadocos/visual-question-answering-study31 Visual Question Answering • 16.6k • Updated Sep 4 • 7
Compile by Training: Turning Natural-Language Specifications into Local Neural Functions Paper • 2609.04199 • Published Sep 3 • 273