Search papers, labs, and topics across Lattice.
5
0
6
20
Training trajectory forecasting models with a metric-agnostic approach can lead to state-of-the-art performance across all evaluation metrics, challenging the notion that metric-specific optimization is necessary.
SurGe reveals that enhancing local surface geometry can significantly elevate the performance of 3D reconstruction models, challenging the status quo of existing evaluation metrics.
Latent dynamics models like Dreamer can lure you into a false sense of security: their epistemic uncertainty estimates are unreliable because they're biased towards high-reward attractors in the latent space, even when the real world is different.
Vanilla Transformers, previously sidelined in 3D scene understanding, now outperform specialized architectures thanks to a clever training recipe that overcomes data scarcity.
Ditch the complex trackers: a plain ViT encoder, augmented with a clever query propagation trick, delivers state-of-the-art video segmentation at 10x the speed.
