Agentic Visualization: Extracting Agent-Based Design Patterns From Visualization Systems
Vaishali Dhanoa, Anton Wolter, Gabriela Molina León, Hans‐Jörg Schulz, Niklas Elmqvist · 2025 · 5 citationsRead the paper
Autonomous agents powered by large language models are transforming artificial intelligence (AI), creating an imperative for the visualization area. However, our field's focus on a human in the sensemaking loop raises critical questions about autonomy, delegation, and coordination for such agentic visualization that preserve human agency while amplifying analytical capabilities. This article addresses these questions by reinterpreting existing visualization systems with semiautomated or fully automatic AI components through an agentic lens. Based on this analysis, we extract a collection of design patterns for agentic visualization, including agentic roles, communication, and coordination. These patterns provide a foundation for future agentic visualization systems that effectively harness AI agents while maintaining human insight and control.
What the full text shows
Theory or simulation- Identification of four primary agent role patterns: Forager (data search), Analyst (sensemaking guidance), Chart Creator (visualization authoring), and Data Steward (data management). (Based on a comprehensive analysis of academic visualization literature)
- Identification of coordination patterns including Scouting (initial independent exploration), Swarming (dividing a single task among multiple agents), and Monitoring (tracking streaming data and notifying users). (Based on a comprehensive analysis of academic visualization literature)
- Identification of communication patterns such as Progress Indicators (status updates) and Provenance Logs (tracking interactions). (Based on a comprehensive analysis of academic visualization literature)
- Setup
- Analysis of existing visualization systems (both pre-LLM and LLM-based) meeting specific criteria: human interaction, data analysis focus, interactive visualization components, and autonomous elements.
- How
- The authors conducted a systematic review of academic visualization literature, applying four inclusion criteria to identify systems with autonomous or semi-autonomous components. They reinterpreted these systems through an agentic lens to extract recurring design patterns appearing in at least two distinct systems.
- Limitations
- Restricted to academic publications for the survey · Patterns are extracted from existing systems rather than validated through a new empirical study
Summarized by Gemma 4 from the open-access full text · 9 Oct 2026. Read it