Ying Shan

WorldCrafter is a video world model that learns a camera-queryable implicit 3D-aware memory that enables streaming scene exploration from a single input image or text prompt and shows substantial gains in long-horizon consistency and camera-control accuracy while preserving visual quality during minute-scale exploration.

The GameHorizon Suite, a unified data and evaluation suite that measures gameplay capabilities at different horizons for diverse model families, can provide a standardized yardstick for evaluating gameplay capabilities across horizons and model families.

Early decision-making in multimodal reasoning can cut inference time without sacrificing accuracy, thanks to a novel dual evaluation of model competencies.

MeshWeaver achieves a groundbreaking 18% compression ratio while generating high-fidelity meshes with over 16K faces by weaving geometry directly into the autoregressive process.