AI Research 技术调研

标签: goal-conditioned

此标签下有5条笔记。

  • 2026年7月16日

    BridgeData V2: A Dataset for Robot Learning at Scale

    • dataset
    • robot-learning
    • teleoperation-data
    • widowx250
    • multi-task
    • language-conditioned
    • goal-conditioned
    • imitation-learning
    • offline-rl
    • cross-institution-generalization
  • 2026年7月16日

    RoboCat: A Self-Improving Foundation Agent for Robotic Manipulation

    • robocat
    • self-improvement
    • goal-conditioned
    • gato
    • vq-gan
    • decision-transformer
    • multi-embodiment
    • cross-embodiment
    • hindsight-goals
    • foundation-agent
    • manipulation
  • 2026年7月16日

    SkillBlender: Towards Versatile Humanoid Whole-Body Loco-Manipulation via Skill Blending

    • humanoid
    • whole-body-control
    • loco-manipulation
    • hierarchical-rl
    • skill-blending
    • goal-conditioned
    • cross-embodiment
    • benchmark
    • reward-hacking
    • isaac-gym
  • 2026年7月16日

    Deep Hierarchical Planning from Pixels (Director)

    • world-model
    • hierarchical-rl
    • feudal-rl
    • model-based-rl
    • rssm
    • goal-conditioned
    • latent-imagination
    • sparse-reward
    • quadruped-navigation
    • atari
  • 2026年7月16日

    Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models

    • jepa
    • world-model
    • offline-rl
    • planning
    • mpc
    • latent-dynamics
    • goal-conditioned
    • trajectory-stitching
    • lecun
    • vicreg

  • GitHub