A transformer with prediction-correction and hierarchical super-token merging unifies simulation of six physical dynamics categories on Lagrangian particles and generalizes to unseen conditions.
International conference on machine learning , pages=
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
representative citing papers
EyeCue detects driver cognitive distraction by modeling gaze-visual context interactions in egocentric videos and achieves 74.38% accuracy on the new CogDrive dataset, outperforming 11 baselines.
A mean-pool deep set trained on sets of size at most two produces an encoder that generalizes to arbitrary sizes, decoupling representation learning from posterior modeling and making training cost independent of deployment set size N.
Video Prediction Policy conditions robot action learning on future-frame predictions inside fine-tuned video diffusion models, yielding 18.6% relative gains on Calvin ABC-D and 31.6% higher real-world success rates.
A self-supervised method learns a fixed set of disentangled fingerprint tokens from medical time series by combining reconstruction loss with a total coding rate diversity penalty, framed as a disentangled rate-distortion problem.