In vision-language models (VLMs), visual tokens usually consume a significant amount of computational overhead, despite their sparser information density compared to text tokens. To address this, ...
Researchers at Los Alamos National Laboratory have developed a new tool called the Prelim Attention Score, or PAS, to help detect when a vision-language model’s output is grounded in the image and ...
Abstract: Indoor scene recognition is a crucial component in vision-and-language navigation (VLN), which involves guiding an agent to navigate through unseen, photo-realistic environments using ...
MRATNet: Learning Discriminative Features for Partial Discharge Pattern Recognition via Transformers
Abstract: Partial discharge pattern recognition (PDPR) is the fundamental cornerstone for fault diagnosis. It has emerged as a pivotal focal point in the field of power systems. However, PDPR faces ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results