Jan Kautz

Latest

• Fast-ThinkAct: Efficient Vision-Language-Action Reasoning via Verbalizable Latent Planning
• OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM
• EoRA: Fine-tuning-free Compensation for Compressed LLM with Eigenspace Low-Rank Approximation
• Mosaic3D: Foundation Dataset and Model for Open-Vocabulary 3D Segmentation
• Hymba: A Hybrid-head Architecture for Small Language Models