ProtoAct: Turning Wet-Lab Protocols into Embodied Robotic Actions

2026-08-03

AI Summary

This work addresses the challenge that biological wet-lab protocols are typically unstructured, human-oriented texts laden with contextual dependencies and implicit state conditions, rendering them unsuitable for direct robotic execution. To bridge this gap, the authors propose ProtoAct, a novel framework that establishes the first end-to-end pipeline translating free-form textual protocols into embodied robot actions. ProtoAct employs ProtoRAG for context-sensitive parsing, RefineChecker to detect and rectify logical inconsistencies or omissions, and ActSchema to generate constraint-aware JSON action sequences. The framework leverages the BioP2E dataset to train Vision-Language-Action (VLA) models. Experiments across seven large language models demonstrate that ProtoAct enables robust, generalizable protocol execution in both simulated and real-world robotic environments, significantly advancing the automation of biological experimentation.

Abstract

Biological wet-lab protocols are written for trained researchers and often leave routine operations, state-dependent conditions, and contextual parameters implicit, making them difficult to translate into robot-executable actions. We present ProtoAct, a structured protocol-grounding framework that converts free-form biological procedures into state-aware, embodiment-ready action sequences. ProtoAct uses ProtoRAG to retrieve manually annotated examples for context-sensitive parsing, employs RefineChecker to detect and revise missing or inconsistent steps, and applies ActSchema to map the refined procedure into constrained JSON function sequences. We further introduce BioP2E, for which we manually annotate 22 cell-culture protocols into 258 monitoring conditions, 910 executable subtasks, and 962 grounded action calls. Evaluation across seven large language models demonstrates that ProtoAct can be effectively instantiated with different backbones. Ablations confirm that retrieval, posterior checking, and schema constraints make complementary contributions. The parsed subtasks further support demonstration collection and VLA model training, enabling successful execution in both simulation and real-robot settings. ProtoAct thus provides a practical interface between biological protocol understanding and embodied robotic execution.

Problem

• wet-lab protocols
• robotic execution
• protocol grounding
• embodied actions
• action translation

Innovation

• ProtoAct
• protocol grounding
• embodied robotic execution
• wet-lab automation
• action schema