AI Research 技术调研

标签: vision-transformer

此标签下有6条笔记。

  • 2026年7月16日

    Where are we in the search for an Artificial Visual Cortex for Embodied Intelligence?

    • vc-1
    • cortexbench
    • mae
    • vision-transformer
    • pretrained-visual-representations
    • embodied-ai
    • egocentric-video
    • ego4d
    • scaling-hypothesis
  • 2026年7月16日

    Render and Diffuse: Aligning Image and Action Spaces for Diffusion-based Behaviour Cloning

    • behavior-cloning
    • diffusion-policy
    • action-diffusion
    • rendered-action-representation
    • spatial-generalization
    • sample-efficiency
    • rlbench
    • vision-transformer
    • ddim
    • gripper-rendering
  • 2026年7月16日

    A-JEPA: Joint-Embedding Predictive Architecture Can Listen

    • jepa
    • audio
    • self-supervised
    • representation-learning
    • spectrogram
    • masking
    • curriculum-learning
    • vision-transformer
  • 2026年7月16日

    Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture (I-JEPA)

    • jepa
    • self-supervised
    • world-model
    • vision-transformer
    • latent-prediction
    • non-generative
    • masking
    • representation-learning
  • 2026年7月16日

    Learning and Leveraging World Models in Visual Representation Learning

    • jepa
    • world-model
    • self-supervised-learning
    • vision-transformer
    • equivariant-representation
    • predictor-finetuning
    • latent-prediction
    • representation-learning
  • 2026年7月16日

    Revisiting Feature Prediction for Learning Visual Representations from Video (V-JEPA)

    • jepa
    • world-model
    • self-supervised
    • video-representation
    • latent-prediction
    • non-generative
    • masking
    • vision-transformer
    • frozen-evaluation
    • lecun
    • 样本

  • GitHub