CLIP Behaves like a Bag-of-Words Model Cross-modally but not Uni-modally Paper • 2502.03566 • Published 21 days ago • 2
ONEBench to Test Them All: Sample-Level Benchmarking Over Open-Ended Capabilities Paper • 2412.06745 • Published Dec 9, 2024 • 6
No "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance Paper • 2404.04125 • Published Apr 4, 2024 • 28