Q-GeoMem: Question-Guided Geometric Memory for Video Spatial Reasoning Paper • 2605.27318 • Published May 26 • 1
view post Post 161 EO-1 is now available directly in LeRobot!You can train, evaluate, and deploy EO-1 through the standard LeRobot policy interface, making it much easier to try EO-1 on robot-control workflows.Huge thanks to @pepijn223 and the lerobot team for the collaboration and support.EO-1 is an open-source Vision-Language-Action model for general robot control. It combines a Qwen2.5-VL backbone with continuous flow-matching, enabling multimodal perception, embodied reasoning, and action generation in one unified model.Try it out:LeRobot: https://github.com/huggingface/lerobot/tree/main/src/lerobot/policies/eo1Project: http://eo-robotics.ai/eo-1Paper: https://arxiv.org/abs/2508.21112Code: https://github.com/SHAILAB-IPEC/EO1 See translation 🚀 3 3 + Reply
DrivingGen: A Comprehensive Benchmark for Generative Video World Models in Autonomous Driving Paper • 2601.01528 • Published Jan 4 • 19