LateOn-Code π» Collection State-of-the-art late interaction code retrieval models β’ 6 items β’ Updated 7 days ago β’ 12
view article Article LightOnOCR-1B: The Case for End-to-End and Efficient Domain-Specific Vision-Language Models for OCR Oct 23, 2025 β’ 73
SmolDocling datasets Collection Datasets used to train SmolDocling β’ 6 items β’ Updated Jul 31, 2025 β’ 31
PrimitiveAnything: Human-Crafted 3D Primitive Assembly Generation with Auto-Regressive Transformer Paper β’ 2505.04622 β’ Published May 7, 2025 β’ 27
Exploring the Limit of Outcome Reward for Learning Mathematical Reasoning Paper β’ 2502.06781 β’ Published Feb 10, 2025 β’ 58
Recurrent Models Collection These are checkpoints for recurrent LLMs developed to scale test-time compute by recurring in latent space. β’ 15 items β’ Updated May 21, 2025 β’ 11