technique
Feature Auto-Encoder
techniqueactiveprovisional
feature-auto-encoder-136ac3d4·1 events·first seen 14d agoAliases: Feature Auto-Encoder
Co-occurring entities
More like this (12)
Sparse AutoencodersSparse AutoencoderSparse Autoencoders (SAEs)Natural Language Autoencodersmasked autoencodingGraph Neural Network EncoderVAE EncoderVariational Autoencoder (VAE)Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse AutoencodersAutoForestYour UnEmbedding Matrix is Secretly a Feature Lens for Text Embeddingsencoder-only language models
Recent events (1)
Apple researchers propose Feature Auto-Encoder to speed diffusion training via compressed DINOv2 embeddings
Researchers at Apple introduced Feature Auto-Encoder (FAE), a latent diffusion image generator that compresses DINOv2 vision encoder embeddings before learning to denoise them, then expands them back for decoding. The approach achieves comparable image quality to state-of-the-art diffusion models while training roughly 7x faster on ImageNet class-conditional generation. The key insight is that shrinking semantically rich vision embeddings reduces compute during diffusion training without sacrificing the representational benefits of large pretrained encoders.