Large vision-language models often struggle with fine-grained image-text alignment in low-resource settings, leading to mode collapse and reduced output diversity. We address this by applying ...
2025 China Graduate IC Innovation Competition · EDA Elite Challenge (Track 2: Circuit System Block Diagram Recognition & Parsing). Before running, make sure the base model + adapters, images, and ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results