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

Crystalis：用于协同多视图可视化生成的渐进式成核与语义退火框架

Abstract: 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.

摘要： 大型语言模型（LLMs）虽然能够生成独立的图表，但对于协同多视图可视化（CMVs）——即视图间共享数据流和跨视图交互的可视化形式——仍难以实现。数据转换、视觉编码和交互协调之间紧密的字段级耦合，导致其中一个组件的错误会悄无声息地使其他组件失效。

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?

与其追求依赖于模型能力、领域知识和用户专业水平的端到端分析质量，我们转而关注一个基础性问题：LLMs 能否可靠地生成结构正确的 CMVs？什么样的抽象机制能使之成为可能？

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).

我们提出了 Crystalis，这是一个基于以查询为中心的 CMV 建模框架。它将 CMV 分解为跨越三种组件类型（数据、可视化、交互）和三个抽象层级（需求、规范、可执行对象）的依赖图上的结构化查询。

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.

两种互补机制作用于该结构：渐进式成核（progressive nucleation）沿着依赖顺序，将每个查询从需求垂直“结晶”为对象；而语义退火（semantic annealing）则通过分层逻辑检查，确保各层级查询之间的水平一致性。

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.

在涵盖五种前沿 LLMs 的 12 项任务基准测试中，Crystalis 实现了高达 75% 的端到端成功率，显著优于代理编码基准（使用相同基础模型时仅为 8.3%）。此外，一项针对 12 位从业者的用户研究证实了该分解与迭代优化工作流的可用性。