# Crystalis: Progressive Nucleation and Semantic Annealing for Coordinated Multi-View Visualization Generation

arXiv: https://arxiv.org/abs/2607.24766 · HTML: https://mlsanity.com/paper/2607.24766/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 54 references that are in the map.
- First posted on arXiv 7 Jun 2026.

## Builds on (most cited first, 16 of 16)

- [NL4DV: A Toolkit for Generating Analytic Specifications for Data Visualization from Natural Language Queries](https://mlsanity.com/paper/2008.10723/) (2020)
- [ChartGPT: Leveraging LLMs to Generate Charts from Abstract Natural Language](https://mlsanity.com/paper/2311.01920/) (2023)
- [Data2Vis: Automatic Generation of Data Visualizations Using Sequence to Sequence Recurrent Neural Networks](https://mlsanity.com/paper/1804.03126/) (2018)
- [VizML: A Machine Learning Approach to Visualization Recommendation](https://mlsanity.com/paper/1808.04819/) (2018)
- [Collecting and Characterizing Natural Language Utterances for Specifying Data Visualizations](https://mlsanity.com/paper/2110.00680/) (2021)
- [Calliope: Automatic Visual Data Story Generation from a Spreadsheet](https://mlsanity.com/paper/2010.09975/) (2020)
- [FlowSense: A Natural Language Interface for Visual Data Exploration within a Dataflow System](https://mlsanity.com/paper/1908.00681/) (2019)
- [LightVA: Lightweight Visual Analytics with LLM Agent-Based Task Planning and Execution](https://mlsanity.com/paper/2411.05651/) (2024)
- [KG4Vis: A Knowledge Graph-Based Approach for Visualization Recommendation](https://mlsanity.com/paper/2107.12548/) (2021)
- [Snowy: Recommending Utterances for Conversational Visual Analysis](https://mlsanity.com/paper/2110.04323/) (2021)
- [MultiVision: Designing Analytical Dashboards with Deep Learning Based Recommendation](https://mlsanity.com/paper/2107.07823/) (2021)
- [DashBot: Insight-Driven Dashboard Generation Based on Deep Reinforcement Learning](https://mlsanity.com/paper/2208.01232/) (2022)
- [Composition and Configuration Patterns in Multiple-View Visualizations](https://mlsanity.com/paper/2007.15407/) (2020)
- [MEDLEY: Intent-based Recommendations to Support Dashboard Composition](https://mlsanity.com/paper/2208.03175/) (2022)
- [VAID: Indexing View Designs in Visual Analytics System](https://mlsanity.com/paper/2211.02567/) (2022)
- [DIEL: Interactive Visualization Beyond the Here and Now](https://mlsanity.com/paper/1907.00062/) (2019)

## Abstract (arXiv)

Large language models (LLMs) can generate individual charts, but coordinated multi-view visualizations (CMVs), where views share data flows and cross-view interactions, remain out of reach. Tight field-level coupling among data transformations, visual encodings, and interaction coordinations causes errors in one component to silently invalidate others. Rather than pursuing end-to-end analytical quality, which depends on model capability, domain knowledge, and user expertise, we target a foundational question: can LLMs reliably produce structurally correct CMVs, and what abstractions make this possible? We present Crystalis, a framework built on query-centric CMV modeling that decomposes a CMV into structured queries over a dependency graph spanning three component types (Data, Visualization, Interaction) and three abstraction levels (requirement, specification, executable object). Two complementary mechanisms operate over this structure: progressive nucleation crystallizes each query vertically from requirement to object along the dependency order, while semantic annealing enforces horizontal consistency across queries at each level through layered logical checks. On a 12-task benchmark across five frontier LLMs, Crystalis achieves up to 75% end-to-end success, substantially outperforming an agentic coding baseline (8.3% E2E with the same foundation model), and a user study with 12 practitioners confirms the usability of the decomposition and iterative refinement workflow.