imagenet
ImageNet is a large-scale visual database widely used for training and benchmarking deep learning models, particularly in the field of image recognition and computer vision.
- Adaptive Discretization for Consistency Models
- Boosting Adversarial Transferability with Spatial Adversarial Alignment
- Bridging the gap to real-world language-grounded visual concept learning
- Efficient and Generalizable Mixed-Precision Quantization via Topological Entropy
- Energy Matching: Unifying Flow Matching and Energy-Based Models for Generative Modeling
- FACE: Faithful Automatic Concept Extraction
- GIST: Greedy Independent Set Thresholding for Max-Min Diversification with Submodular Utility
- How Classifier Features Transfer to Downstream: An Asymptotic Analysis in a Two-Layer Model
- Knowledge Distillation Detection for Open-weights Models
- OOD Detection with Relative Angles
- Representation Entanglement for Generation: Training Diffusion Transformers Is Much Easier Than You Think
- Representation Entanglement for Generation: Training Diffusion Transformers Is Much Easier Than You Think
- Representational Difference Explanations
- Sparse Image Synthesis via Joint Latent and RoI Flow
- SparseDiT: Token Sparsification for Efficient Diffusion Transformer
- Spiking Neural Networks Need High-Frequency Information
- Training the Untrainable: Introducing Inductive Bias via Representational Alignment
- TransferBench: Benchmarking Ensemble-based Black-box Transfer Attacks