Search papers, labs, and topics across Lattice.
University of Illinois Chicago
5
0
7
UQ rankings for GUI grounding are stable within model classes but falter across different models and interfaces, highlighting the need for tailored calibration in practical applications.
GroundControl reveals that anticipating navigation failures through trajectory-consistent uncertainty can drastically improve the reliability of vision-language agents in real-world applications.
VLMs can ace the ranking but bomb the scoring, revealing a critical flaw in how we evaluate multimodal systems.
Decomposing uncertainty into aleatoric and epistemic types lets robots recover from errors 21% more effectively than treating all uncertainty the same.
Transformer-based visual trackers can slash compute by up to 12% without sacrificing accuracy, simply by dynamically adjusting their depth based on uncertainty.
