Search papers, labs, and topics across Lattice.
The University of Tokyo
3
0
6
Finally, realistic head movements that naturally complement gaze can be automatically generated from video, thanks to a new data-driven approach.
LLMs can now learn mathematical reasoning 2x faster and with greater stability, thanks to a new token-level policy optimization method.
Forget hand-tuning: PolyFormer learns to automatically simplify complex, physically-constrained optimization problems into efficient polytopic reformulations, achieving massive speedups and memory reductions.
