Embodied AI 101

Shaoqing Tan
  • 0,0 (0)
  • Tecnología
  • Diario

Stay in the loop on research in AI and physical intelligence.

  1. hace 4 d

    Imagining Locomotion: Learning a Neural World Model for Legged Robots

    A neural dynamics world model paired with model-free RL policies for quadruped and humanoid locomotion in IsaacLab, enabling long-horizon autoregressive prediction and imagined rollouts that outperform pure model-based RL in prediction accuracy, policy learning, and sim-to-real transfer.

  2. hace 4 d

    One Model to See, Plan, and Act: Introducing EO-1 for Embodied AI

    A 3B parameter unified decoder-only transformer that interleaves vision, text, and action tokens for perception, planning, reasoning, and control in a single model. Trained on the 1.5M-sample EO-1.5M dataset with strong results across manipulation tasks and benchmarks.

  3. hace 4 d

    Video-Action Models for Robot Learning

    Introduces Video-Action Models (VAMs) that leverage pretrained internet-scale video models such as Cosmos-Predict2 as backbones instead of VLMs, paired with a flow-matching action decoder. Claims approximately 10x sample efficiency gains over standard vision-language-action models on real-world pick-and-place tasks.

  4. hace 4 d

    FastDSAC: Unlocking the Potential of Maximum Entropy RL in High-Dimensional Humanoid Control

    Introduces FastDSAC, a method that scales maximum entropy reinforcement learning to high-dimensional humanoid control tasks, improving sample efficiency and policy performance.

  5. hace 4 d

    A Minimalist Retargeting-Guided Reinforcement Learning Recipe for Dexterous Manipulation

    Proposes a minimalist pipeline using retargeting-guided RL for dexterous manipulation in humanoid robots. Focuses on simplifying the training recipe while maintaining strong manipulation performance.

  6. hace 5 d

    DexNDM: Closing the Reality Gap for Dexterous In-Hand Rotation

    DexNDM bridges the sim-to-real gap for stable in-hand rotation of complex objects, enabling learning from biased real-world data without requiring any successful demonstrations.

  7. hace 5 d

    Learning Multi-Modal Trajectory Policies for Data-Efficient Robotic Manipulation

    Proposes learning multi-modal trajectory policies to improve data efficiency in robotic manipulation tasks.

  8. hace 6 d

    ACE: Agentic Control for Embodied Manipulation via Zero-shot Workflow Reasoning

    A zero-shot workflow reasoning approach for agentic control of embodied manipulation, enabling robots to perform complex tasks without task-specific training.

Ver todo (230)

Información

Stay in the loop on research in AI and physical intelligence.

Ficha técnica

  • Creado por
    Shaoqing Tan
  • Años en activo
    2025 - 2026
  • Episodios
    230
  • Clasificación
    Apto
  • Derechos de autor
    © 2026 Shaoqing Tan

Quizá también te guste