Search papers, labs, and topics across Lattice.
University of South Florida
3
0
8
Even top-performing Vision Language Models miss 30% of logical anomalies, revealing critical gaps in their reasoning abilities.
LLMs leak training data membership through subtle, gradient-induced shifts in their internal representations, enabling a surprisingly effective new attack.
Autonomous underwater vehicles can now navigate complex environments more effectively by combining the reasoning of vision-language models with dynamics-aware planning, achieving better coverage and zero collisions.
