Search papers, labs, and topics across Lattice.
6
0
9
5
Achieving 78% success in real-world grasping, SeededGrasp redefines efficiency in robotic manipulation by decoupling semantic reasoning from geometric execution.
Color bias from black-level errors can severely impact low-light image denoising, but a new calibration-free approach effectively corrects this issue, outperforming traditional methods.
Forget fixed agent slots and quadratic attention: Gamma-World uses simplex embeddings and sparse hubs to generate interactive multi-agent environments with better fidelity and control, even generalizing from 2 to 4 players without retraining.
Achieve subject-driven image generation that actually understands both text prompts and subject identity by squeezing more capacity out of multimodal LLMs.
You can slash the compute cost of visual geometry transformers by 85% without sacrificing accuracy by intelligently pruning redundant tokens across frames and within layers.
Training data is not enough: reasoning traces from diverse cultural backgrounds are critical for safe and reliable autonomous driving in rare, long-tail scenarios.
