A2P-Vis: an Analyzer-to-Presenter Agentic Pipeline for Visual Insights Generation and Reporting

2025-12-26

AI Summary

Automated end-to-end data science suffers from “chart-rich but insight-scarce, rigid reporting” bottlenecks. Method: This paper proposes a two-stage multi-agent pipeline: the Analyzer agent performs data profiling, generates diverse visualizations, conducts readability validation (via hybrid rule-based and LLM-driven checks), and scores insights across dimensions—depth, correctness, and actionability; the Presenter agent organizes themes, plans narrative structure, and refines documentation into coherent, professional reports. Contribution/Results: We introduce the first analyzer–presenter collaborative paradigm that unifies automated insight quality assessment with narrative-level report generation. The system integrates code-generation-and-execution agents, hybrid readability validation, a multidimensional insight scoring model, and a narrative planning agent. Evaluated on multiple real-world datasets, it significantly outperforms single-stage baselines and enables out-of-the-box generation of publication-ready visualization reports.

Technology Category

• Data Mining & Knowledge Management: Data Visualization & Summarization
• Multiagent Systems: Modeling other Agents
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Web Mining and Content Analysis: Web data visualization
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Automating end-to-end data science pipeline with AI agents still stalls on two gaps: generating insightful, diverse visual evidence and assembling it into a coherent, professional report. We present A2P-Vis, a two-part, multi-agent pipeline that turns raw datasets into a high-quality data-visualization report. The Data Analyzer orchestrates profiling, proposes diverse visualization directions, generates and executes plotting code, filters low-quality figures with a legibility checker, and elicits candidate insights that are automatically scored for depth, correctness, specificity, depth and actionability. The Presenter then orders topics, composes chart-grounded narratives from the top-ranked insights, writes justified transitions, and revises the document for clarity and consistency, yielding a coherent, publication-ready report. Together, these agents convert raw data into curated materials (charts + vetted insights) and into a readable narrative without manual glue work. We claim that by coupling a quality-assured Analyzer with a narrative Presenter, A2P-Vis operationalizes co-analysis end-to-end, improving the real-world usefulness of automated data analysis for practitioners. For the complete dataset report, please see: https://www.visagent.org/api/output/f2a3486d-2c3b-4825-98d4-5af25a819f56.

Problem

• Automates end-to-end data science pipeline with AI agents
• Generates insightful visual evidence and coherent professional reports
• Converts raw datasets into publication-ready data-visualization narratives

Innovation

• Two-part multi-agent pipeline for automated data visualization reports
• Analyzer generates and filters visualizations with insight scoring
• Presenter composes narrative reports from curated charts and insights