Towards Agentic Robots: Model, Data, Environment, and Harness

Key points

• Foundation models are driving a paradigm shift in robotic intelligence, moving beyond low-level action prediction to goal-directed orchestration of data collection, model training, task planning, and real-world execution.1
• This paper provides a unified overview of existing research on agentic robots.1
• We first trace the development from robot policies to agentic systems, covering vision-language-action models, world models, and world-action models.1
• Across these domains, we examine how agents shape model learning and robot decisions, collect and curate robot data, construct worlds, tasks, and tests for learning and evaluation, and organize capabilities for robot use.1

Sources (2)

• [1]Towards Agentic Robots: Model, Data, Environment, and Harness
  Hugging Face Daily Papers · Oct 14, 12:00 AM

  We characterize this emerging paradigm as an agentic robots, in which agents reason over human instructions, interact with robotic environments, rectify execution errors autonomously, and dynamically adjust operational strategies.

  Foundation models are driving a paradigm shift in robotic intelligence, moving beyond low-level action prediction to goal-directed orchestration of data collection, model training, task planning, and real-world execution.

• [2]Towards Agentic Robots: Model, Data, Environment, and Harness
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 14, 04:32 PM · same content

Extractive summary: sentences quoted from the sources.