Search papers, labs, and topics across Lattice.
Eyeline Labs
5
0
7
5
ID-V2V enables seamless video edits while preserving facial identity and performance, revolutionizing the way we approach generative video models.
Achieving seamless video compositing and motion control in a single model could revolutionize filmmaking techniques and workflows.
ProfiLLM achieves a remarkable 6.14% boost in prediction accuracy for ride-hailing dispatch by transforming user behavior into actionable profiles using LLMs.
Achieving multi-view consistency in 3D face re-aging, ReAge3D preserves subtle age-related details that traditional methods overlook.
Forget painstakingly collecting real-world data for deformable object manipulation: SIM1 achieves equivalent policy performance with 15x less data by grounding simulations in the physical world.
