EmbodiedClaw: Conversational Workflow Execution for Embodied AI Development

arXiv cs.RO / 4/16/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

EmbodiedClaw proposes a new embodied AI development paradigm where researchers specify goals and constraints via conversation, and the system then plans and runs the workflow automatically.
The approach targets the large engineering overhead of embodied AI’s multi-task, multi-scene, and multi-model setup stages, including evaluation environment creation, trajectory collection, training, and evaluation.
EmbodiedClaw is implemented as a conversational agent that converts expensive, high-frequency research activities (e.g., environment creation/revision, benchmark transformation, trajectory synthesis, model evaluation, and asset expansion) into executable skills.
Experimental results across end-to-end workflow tasks, capability-specific evaluations, human researcher studies, and ablations indicate reduced manual engineering effort alongside gains in executability, consistency, and reproducibility.

Abstract

Embodied AI research is increasingly moving beyond single-task, single-environment policy learning toward multi-task, multi-scene, and multi-model settings. This shift substantially increases the engineering overhead and development time required for stages such as evaluation environment construction, trajectory collection, model training, and evaluation. To address this challenge, we propose a new paradigm for embodied AI development in which users express goals and constraints through conversation, and the system automatically plans and executes the development workflow. We instantiate this paradigm with EmbodiedClaw, a conversational agent that turns high-frequency, high-cost embodied research activities, including environment creation and revision, benchmark transformation, trajectory synthesis, model evaluation, and asset expansion, into executable skills. Experiments on end-to-end workflow tasks, capability-specific evaluations, human researcher studies, and ablations show that EmbodiedClaw reduces manual engineering effort while improving executability, consistency, and reproducibility. These results suggest a shift from manual toolchains to conversationally executable workflows for embodied AI development.