## Notes on spatial intelligence

### News

**LeRobot goes to driving school**  
Yaak and Hugging Face are teaming up to create the world’s largest open-source, self-driving dataset: Learning to Drive (L2D).  
[Read more](https://www.yaak.ai/blog/lerobot-goes-to-driving-school)

**Building Spatial Intelligence - Part 2**  
While working with petabyte scale datasets from the automotive domain, we identified 4 key areas essential to building end-to-end (e2e) AI models for robotics: Logs, Discovery, Multimodal data, and Spatial intelligence.  
[Read more](https://www.yaak.ai/blog/building-spatial-intelligence-part-2)

**Building Spatial Intelligence - Part 1**  
Movement is essential to intelligence. The cycle of perceiving and acting is key to our understanding of the physical world around us: cleaning a tabletop, navigating an intersection, or moving household items. These seemingly trivial tasks hide a complex choreography of sensory-motor abilities in plain sight.  
[Read more](https://www.yaak.ai/blog/buildling-spatial-intelligence-part1)

**Next action prediction with GPTs**  
In this blog, we extrapolate the idea of next-word prediction (given a partial sentence) to next-action prediction (given the context, e.g. image, sensor data). We found that a simple formulation doesn't work out of the box but rather requires careful understanding of state-action spaces and causality.  
[Read more](https://www.yaak.ai/blog/next-action-prediction-with-gpts)

**A novel test for autonomy**  
We shed light on learning to drive and testing for human drivers within the context of current EU regulations. We hypothesize that training lessons and driving tests are a possible blueprint for a test of generalization for …  
[Read more](https://www.yaak.ai/blog/a-novel-test-for-autonomy)