Towards Reliable Agentic Progressive Text-to-Visualization with Verification Rules

2026-05-28

AI Summary

Existing text-to-visualization approaches predominantly rely on a single-round, one-shot generation paradigm, which often leads to cognitive overload and visualization errors. To address this limitation, this work proposes PMVis—a progressive, multi-turn interactive framework that enables users to iteratively refine their intent through dialogue. The approach introduces PMVisBench, the first benchmark dataset tailored for this task, and PMVisAgent, a multi-agent system incorporating validation and repair mechanisms. By integrating visual query language (VQL) simplification, natural language query (NLQ) reconstruction, ReAct-style tool invocation, and explicit interaction-based verification rules, PMVis significantly outperforms existing methods, achieving accuracy improvements of 17.57% and 23.21% on single-table and multi-table scenarios, respectively.

Abstract

Text-to-Visualization (Text-to-Vis) translates natural language queries into visualization query languages, enabling non-expert users to perform data analysis. However, most existing methods follow a one-shot paradigm that requires users to specify all visualization details in a single round, often leading to cognitive overload and incorrect visualizations. In this paper, we propose PMVis, a progressive multi-turn paradigm for text-to-vis, where users' intents are refined through multi-turn interactions. To support research in this paradigm, we construct PMVisBench, the first dataset designed to capture the progressive and iterative nature of real-world user queries. It is built through VQL simplification and NLQ reconstruction, with explicit rule constraints to ensure each intermediate VQL remains valid and meaningful. Building upon PMVis, we further introduce PMVisAgent, an agent-based framework that simulates realistic user-system dialogues. PMVisAgent consists of a User, a System, and a Validation Agent that performs verification and repair via a ReAct-style tool-use loop to mitigate error accumulation across rounds, with explicit interaction and verification rules to ensure reliability of the multi-agent system. Extensive experiments on PMVisBench demonstrate that PMVisAgent significantly outperforms state-of-the-art text-to-vis baselines. It achieves up to 17.57\% and 23.21\% improvements in execution accuracy in single-table and multi-table settings, respectively, while ablation studies confirm the importance of combining progressive interaction with clarification. The code is available at https://github.com/wxxv/PMVis.

Problem

• Text-to-Visualization
• one-shot paradigm
• cognitive overload
• incorrect visualizations
• user intent

Innovation

• Progressive Text-to-Visualization
• Multi-turn Interaction
• Verification Rules
• Agent-based Framework
• PMVisBench