Search papers, labs, and topics across Lattice.
4
0
8
Iterative reinforcement learning with human feedback transforms robot gestures from stiff and unnatural to fluid and expressive, revolutionizing human-robot communication.
Unlock zero-shot human motion understanding across diverse wearable IMU setups with AnyMo, a geometry-aware framework that bridges the gap between synthetic data and real-world deployment.
Continuous-depth transformers, augmented with physics-informed loss, can significantly improve short-term weather forecasting, suggesting a promising path for hybrid physics-aware AI models.
General-purpose LLMs stumble on domain-specific RAG, but a model trained on just 6.5K synthetic examples closes the gap and slashes hallucinations almost in half.
