DeepVIS: Bridging Natural Language and Data Visualization Through Step-wise Reasoning

2025-08-03

AI Summary

Existing NL2VIS methods lack transparent reasoning mechanisms, hindering user understanding and correction of generated visualizations. To address this, we propose a Chain-of-Thought (CoT)-guided reasoning framework for natural language-to-visualization (NL2VIS), construct nvBench-CoT—the first NL2VIS benchmark with fine-grained, human-annotated reasoning traces—and design DeepVIS, an interactive system enabling real-time user correction. Technically, our approach integrates CoT-informed large language model fine-tuning, an automated pipeline for reasoning trace annotation, and an interpretable frontend interface. Experiments demonstrate significant improvements over baselines in visualization quality (BLEU-4 +12.3%, F1 +9.7%) and user controllability. A user study confirms high interpretability of the CoT reasoning process and shows that interactive correction substantially increases final visualization satisfaction (+38.5%).

Technology Category

• Computer Vision: Visual Reasoning & Symbolic Representations
• Natural Language Processing: Safety and Robustness
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Web Mining and Content Analysis: Web data visualization
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Although data visualization is powerful for revealing patterns and communicating insights, creating effective visualizations requires familiarity with authoring tools and often disrupts the analysis flow. While large language models show promise for automatically converting analysis intent into visualizations, existing methods function as black boxes without transparent reasoning processes, which prevents users from understanding design rationales and refining suboptimal outputs. To bridge this gap, we propose integrating Chain-of-Thought (CoT) reasoning into the Natural Language to Visualization (NL2VIS) pipeline. First, we design a comprehensive CoT reasoning process for NL2VIS and develop an automatic pipeline to equip existing datasets with structured reasoning steps. Second, we introduce nvBench-CoT, a specialized dataset capturing detailed step-by-step reasoning from ambiguous natural language descriptions to finalized visualizations, which enables state-of-the-art performance when used for model fine-tuning. Third, we develop DeepVIS, an interactive visual interface that tightly integrates with the CoT reasoning process, allowing users to inspect reasoning steps, identify errors, and make targeted adjustments to improve visualization outcomes. Quantitative benchmark evaluations, two use cases, and a user study collectively demonstrate that our CoT framework effectively enhances NL2VIS quality while providing insightful reasoning steps to users.

Problem

• Bridging natural language and data visualization with reasoning
• Enhancing transparency in automated visualization design process
• Improving user understanding and refinement of visualization outputs

Innovation

• Integrates Chain-of-Thought reasoning into NL2VIS pipeline
• Introduces nvBench-CoT dataset for detailed reasoning steps
• Develops DeepVIS interface for interactive reasoning inspection