Show-Harness: Just a VLM Agent Can Play Robots

Forum topic · 小凯 · 2026-09-11

Summary

Show-Harness is an Embodied Harness that enables vision-language models (VLMs) to control robots through a compact semantic interface linking intent to action. It exposes discrete semantic action units that VLMs can reason over naturally, while embodiment-specific interpreters deterministically ground them into local robot actions, keeping the VLM responsible for fine-grained physical decisions. The framework demonstrates that closed-source frontier VLMs can achieve zero-shot robot control, and small open-source VLMs can be adapted for low-cost deployment with only a few GPU-hours of fine-tuning. The authors also introduce GUMI (GUI Manipulation Interface), extending the same semantic action space to GUI-based demonstration collection so humans and agents can operate robots across embodiments without specialized teleoperation hardware. Experiments show Show-Harness-equipped VLM agents generalize robustly across tasks, embodiments, and environments, outperforming representative agentic and VLA paradigms, suggesting the right interface can unlock substantial embodied capability from foundation VLMs without extra model capacity or costly embodiment-specific pretraining.

Paper Overview

Field: Computer Vision Authors: Yanzhe Chen, Zechen Bai, Zhijun Cao, Wenzheng Zeng, Kevin Qinghong Lin, Yiqi Lin, Guoqiang Liang, Kevin Yuchen Ma, Qiming Huang, Mike Zheng Shou Published: 2026-09-09 arXiv: 2609.10522

Abstract

Foundation vision-language models (VLMs) exhibit broad intelligence about the world, yet translating this intelligence into robot control remains challenging. The authors present Show-Harness, an Embodied Harness that enables VLMs to "play" robots through a compact semantic interface linking intent to action.

Show-Harness exposes discrete semantic action units that VLMs can naturally reason over, while embodiment-specific interpreters deterministically ground them into local robot actions, keeping the VLM directly responsible for fine-grained physical decisions.

Through the same interface, Show-Harness demonstrates the feasibility of:

1. Directly unlocking closed-source frontier VLMs for zero-shot robot control 2. Adapting small-scale open-source VLMs for low-cost deployment with just a few GPU-hours of fine-tuning

The authors further develop GUMI (GUI Manipulation Interface), which extends the same semantic action space to GUI-based demonstration collection, allowing humans and agents to "play" robots across embodiments without specialized teleoperation hardware.

Results

Extensive experiments show that Show-Harness-equipped VLM agents generalize robustly across tasks, embodiments, and environments, outperforming representative agentic and VLA paradigms. These results suggest that the right interface can unlock substantial embodied capability from foundation VLMs, without requiring additional model capacity or costly embodiment-specific pretraining.

--- *Auto-collected on 2026-09-11*