Optimizing Visual Demonstration Selection for MLLMs
Discover how Learning to Select Demonstrations (LSD) improves visual in-context learning in MLLMs, outperforming kNN in complex regression tasks.
Multimodal Annotation Framework for Broadcast TV Analytics
Explore a multimodal annotation framework enhancing broadcast TV analytics with AI models, improving content classification and audience engagement insight...
Edge Reliability Challenges in Compressed Vision-Language Models
Explore failure modes of compressed vision-language models under visual corruption and their impact on edge deployment reliability.
Chinese Handwriting Assessment Using Vision-Language Models
Leverage vision-language models for advanced Chinese handwriting assessment with multi-level feedback to improve learners' writing skills effectively.
Tiny-ViT: Efficient Potato Leaf Disease Detection Model
Discover Tiny-ViT, a compact Vision Transformer for accurate, fast, and explainable potato leaf disease classification to boost crop health and yield.