Alex Open Research Wiki

Tag: vision

15 items with this tag.

  • Jun 04, 2026

    Gemma 4 12B Encoder-Free Multimodal Release

    • multimodal
    • encoder-free
    • production-models
    • vision
    • audio
  • Jun 04, 2026

    Vision Foundation Models

    • vision
    • foundation-models
  • May 25, 2026

    leAutoencoder

    • entity
    • autoencoders
    • self-supervised-learning
    • jepa
    • vision
  • May 25, 2026

    Self-Teaching Autoencoder

    • self-supervised-learning
    • autoencoders
    • jepa
    • sigreg
    • representation-learning
    • vision
  • May 22, 2026

    Florence-2

    • entity
    • vision
    • vision-language
    • dataset-bootstrapping
  • May 22, 2026

    RAEv2

    • entity
    • vision
    • representation-autoencoders
    • world-models
  • May 22, 2026

    Florence-2: Advancing a Unified Representation for a Variety of Vision Tasks

    • vision
    • vision-language
    • dataset-bootstrapping
    • data-engine
    • unified-representation
  • May 22, 2026

    Improved Baselines with Representation Autoencoders

    • vision
    • representation-autoencoders
    • latent-space
    • generation
    • world-models
  • May 18, 2026

    DINOv3

    • entity
    • vision
  • May 18, 2026

    Perception Encoder

    • entity
    • vision
    • vision-language
  • May 18, 2026

    DINOv3

    • vision
    • self-supervised-learning
  • May 18, 2026

    Next-Embedding Prediction Makes Strong Vision Learners

    • vision
    • predictive-learning
    • self-supervised-learning
  • May 18, 2026

    Perception Encoder

    • vision
    • vision-language
    • representation-learning
    • intermediate-layers
  • May 18, 2026

    The Prism Hypothesis: Harmonizing Semantic And Pixel Representations Via Unified Autoencoding

    • vision
    • spectral-representations
    • autoencoding
  • May 18, 2026

    Tuna-2: Pixel Embeddings Beat Vision Encoders For Multimodal Understanding And Generation

    • multimodal
    • pixel-space
    • vision

Created with Quartz v4.5.2 © 2026