Debugging Defective Visualizations: Empirical Insights Informing a Human-AI Co-Debugging System

TL;DR

This paper investigates the debugging of visualizations, analyzing existing forum and AI approaches, and introduces a hybrid human-AI system that significantly improves debugging success rates.

Contribution

It provides empirical insights into visualization debugging challenges and presents a novel hybrid system combining LLM suggestions with forum knowledge.

Findings

Forum solutions are accurate but slow.

LLMs provide quick guidance but can be inaccurate.

Hybrid system resolves 86% of debugging cases.

Abstract

Visualization authoring is an iterative process requiring users to adjust parameters to achieve desired aesthetics. Due to its complexity, users often create defective visualizations and struggle to fix them. Many seek help on forums (e.g., Stack Overflow), while others turn to AI, yet little is known about the strengths and limitations of these approaches, or how they can be effectively combined. We analyze Vega-Lite debugging cases from Stack Overflow, categorizing question types by askers, evaluating human responses, and assessing AI performance. Guided by these findings, we design a human-AI co-debugging system that combines LLM-generated suggestions with forum knowledge. We evaluated this system in a user study on 36 unresolved problems, comparing it with forum answers and LLM baselines. Our results show that while forum contributors provide accurate but slow solutions and LLMs offer immediate but sometimes misaligned guidance, the hybrid system resolves 86% of cases, higher than either alone.