UncertaintyVis: Preserving Linguistic Uncertainty in Automated Text-to-Chart Generation

2026-08-07

AI Summary

This study addresses a critical limitation in current text-to-visualization systems: their tendency to overlook linguistic markers of uncertainty (e.g., “approximately,” “at least”), thereby producing charts that convey spurious certainty and mislead readers. To remedy this, the authors introduce the first four-category typology of linguistic uncertainty tailored for chart generation, along with a non-intrusive visual encoding strategy that explicitly communicates uncertainty while preserving spatial layout and readability. The proposed end-to-end pipeline integrates large language models for textual analysis, uncertainty classification, and visualization rendering. User studies demonstrate that the approach achieves 85% accuracy in aligning chart semantics with source text intent, with 75% of participants preferring uncertainty-aware visualizations and reporting reduced cognitive load. While bar and pie charts exhibit robust performance, line charts indicate room for further refinement.

Abstract

Data-rich documents pair narrative text with quantitative claims, and authors routinely qualify those claims with linguistic uncertainty markers such as "nearly," "approximately," or "at least." Automated text-to-chart systems discard these markers, producing visualizations that appear definitive even when the source text expresses hedged or incomplete knowledge. Readers may then over-interpret precision and misjudge author intent. We present UncertaintyVis, a system that preserves linguistic uncertainty during automated chart generation. A formative corpus analysis of 211 uncertainty expressions across 12 documents and 8 domains yielded a four-category taxonomy: Surface Form Normalization, Precision Boundaries, Inferential Derivation, and Non-Inferable Gaps. We mapped each category to chart-specific visual encodings that signal uncertainty without disturbing the spatial integrity readers rely on, and implemented an end-to-end pipeline pairing large language model text analysis with uncertainty-aware rendering. In a two-part study with 12 participants, readers matched charts to source text with 85% accuracy and text to charts with 76%. Uncertainty-aware visualizations trended toward lower cognitive demand (effect sizes 0.460 and 0.769 for mental demand and effort), and 75% of participants preferred them to plain text, describing explicit uncertainty encodings as a basis for verifying data claims. Encoding effectiveness varied by chart type: bar and pie encodings performed consistently, while line chart encodings require redesign.

Problem

• linguistic uncertainty
• text-to-chart generation
• uncertainty visualization
• automated visualization
• data communication

Innovation

• linguistic uncertainty
• text-to-chart generation
• uncertainty visualization
• visual encoding
• large language models