Search papers, labs, and topics across Lattice.
4
0
5
0
Current vision-language models fail to achieve embodied self-awareness, with none surpassing a 16.8% success rate in real-world interaction tasks.
Ms.Forcing achieves a 39.6% speedup in streaming video generation while significantly enhancing quality by intelligently adapting spatial granularity to noise levels.
A groundbreaking dataset reveals that 3D particle models significantly outperform 2D video models in capturing the complexities of deformable object dynamics.
Human motion generation gets a dose of reality: IAM shows that explicitly modeling body morphology and identity leads to more realistic and consistent movements.
