ARCS: Towards Precise Text-to-SQL via Structured Disambiguation Paper • 2610.09396 • Published 16 days ago • 13
GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed Paper • 2609.39600 • Published 23 days ago • 58
PolyOCR-Venus: Unified OCR Foundation Models for Text-Centric Visual Intelligence Paper • 2609.37712 • Published 24 days ago • 38
SearchJev: A Fast and Calibrated System-1 Model for Search Agents Paper • 2610.05107 • Published 19 days ago • 41
Language Models that Play Chess and Explain Their Moves Paper • 2610.03695 • Published 21 days ago • 37