Search papers, labs, and topics across Lattice.
Affiliation:
5
0
5
Counterintuitively, a single off-the-shelf earbud IMU tracks 3D lower-body kinematics at 79 mm error, whereas adding commercial insole sensors actually degrades pose reconstruction due to orientation noise.
Video world models can now faithfully respect coarse, user-designed 3D scene proxies without ever training on a single hand-crafted proxy-video pair.
More wearable sensors don't guarantee better motion capture: firmware-fused foot orientation actively destroys kinematic accuracy unless deep networks learn to dynamically gate reliability at the individual channel level.
Standard early-exit architectures collapse completely outside their pre-selected exits, but stochastic prefix supervision transforms a normal Transformer into a valid, deployable language model at literally every intermediate layer depth.
Achieving an overall score of 84.76, DreamX-World 1.0 sets a new benchmark for interactive video generation, outperforming leading models in both camera control and visual fidelity.
