DEEM is an exploration of using diffusion models as the eyes of multi-modal large language models, with the goal of eliminating potential biases in different visual encoders from a vision-centric ...
Abstract: Interpretable image classification is crucial for making decisions in high-stakes scenarios. Recent advancements have demonstrated that interpretable models can achieve performance ...
Abstract: Recently, textual prompt tuning has shown inspirational performance in adapting Contrastive Language-Image Pre-training (CLIP) models to natural image quality assessment. However, such ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback