Search papers, labs, and topics across Lattice.
University of Wisconsin-Madison
8
0
12
6
A single 500-hour visual-tactile dataset boosts zero-shot contact prediction from 13% to 38% IoU and quadruples robot manipulation success rates, proving scale and consistency are paramount for embodied learning.
Robotic agents can now learn complex manipulation strategies from just a few demonstrations by treating them as evidence to be analyzed, not just training data.
AnaDiffusion allows for precise and controllable editing of 3D brain MRIs, achieving the lowest FID scores while preserving anatomical integrity.
Generating missing multi-view data from diverse driving videos boosts closed-loop driving robustness in edge cases by over 30%.
MoRE achieves a staggering 44 percentage point increase in deployment success rates by seamlessly integrating behavior mode redirection into policy weights, eliminating the need for inference-time adjustments.
Current video editing AIs still struggle to balance visual quality, instruction adherence, and localized edits, as revealed by a new benchmark designed to disentangle these factors.
LLM-generated explanations often fail to help users identify incorrect answers, and simply scaling models or applying post-training doesn't fix the problem.
Generating videos from multiple views of a subject is now possible, thanks to a new method that enforces 3D consistency and avoids confusion between different subjects and views.
