ImageNet-10

'ImageNet-10' is a subset of the ImageNet dataset that contains a limited number of classes and is used to evaluate the performance of multimodal systems in vision-language understanding.

Papers using ImageNet-10 (15)

Mitigating Visual Hallucinations in Multimodal Systems through Retrieval-Augmented Reliability-Aware Inference2026Immuno-VLM: Immunizing Large Vision-Language Models via Generative Semantic Antibodies for Open-World Trustworthiness2026JEPA-T: Joint-embedding Predictive Architecture With Text Fusion For Image Generation2025Multimodal Large Language Models as Image Classifiers2026EvoTok: A Unified Image Tokenizer via Residual Latent Evolution for Visual Understanding and Generation2026High-Fidelity Text-to-Image Generation from Pre-Trained Vision-Language Models via Distribution-Conditioned Diffusion Decoding2026Explaining CLIP Zero-shot Predictions Through Concepts2026Koo-Fu CLIP: Closed-Form Adaptation of Vision-Language Models via Fukunaga-Koontz Linear Discriminant Analysis2026A Hidden Semantic Bottleneck in Conditional Embeddings of Diffusion Transformers2026Cross-modal Proxy Evolving for OOD Detection with Vision-Language Models2026Vision Also You Need: Navigating Out-of-Distribution Detection with Multimodal Large Language Model2026Explaining Similarity in Vision-Language Encoders with Weighted Banzhaf Interactions2025ViLU: Learning Vision-Language Uncertainties for Failure Prediction2025Beginning With You: Perceptual-initialization Improves Vision-language Representation And Alignment2025Image Recognition With Vision And Language Embeddings Of Vlms2025