Compositional Generalization Requires Linear, Orthogonal Representations in Vision Embedding Models Paper • 2602.24264 • Published Feb 27 • 14
Enhancing Multi-Image Understanding through Delimiter Token Scaling Paper • 2602.01984 • Published Feb 2 • 5
DISCO: Diversifying Sample Condensation for Efficient Model Evaluation Paper • 2510.07959 • Published Oct 9, 2025 • 15
Diffusion Classifiers Understand Compositionality, but Conditions Apply Paper • 2505.17955 • Published May 23, 2025 • 22
