arXiv · 2611.12112
LiveWorkspace: Externalizing Visual Understanding as Persistent Shared State for Screenshot-to-Code Agents
Abstract
Screenshot-to-Code requires coding agents to sustain visual understanding across code generation, execution, and revision. We introduce LiveWorkspace, a visual coding harness that externalizes this understanding as Shared Visual State (Shared VS), an implementation-independent representation of visual hierarchy, node attributes, and layout relations. LiveWorkspace constructs and retains \textit{Target} from screenshots and recovers \textit{Current} from browser execution in the same schema. Structured Diff and implementation bindings guide code modifications; user edits update \textit{Target} and reuse the alignment loop. Using GPT-5.2 for both methods, LiveWorkspace improves end-to-end Avg by 1.8 points on Design2Code and 2.0 on ScreenBench over a control with matched model-call counts and stagewise budget limits. Further experiments demonstrate reuse of constructed Shared VS across coders and improvements in iterative repair. LiveWorkspace also achieves 93.0\% edit success across 100 controlled editing cases. These findings support treating visual understanding as a persistent shared working state throughout code generation, execution, and modification.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Shijun Xiao, Zonghui Yang, Xinrong Chen, Hao Liu, Wensheng Zhang, Yongqiang Tang. 2026-11-09. LiveWorkspace: Externalizing Visual Understanding as Persistent Shared State for Screenshot-to-Code Agents. https://arxiv.science/abs/2611.12112
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite arXiv Science by Indiciae: BibTeX · RIS.