XAI Gamification 2024
XAI Gamification explores how stories and game characters can help people without a technical background understand AI. The prototype uses large language models to power conversational characters that guide users through visualizations of model predictions and decision boundaries. We use these interactions to investigate how AI explanations can connect more closely to a person's interests and experience.
Recognition & Outreach
- Accepted to CHI 2025 as a Late-Breaking Work paper.
Core Features
- Talk with AI-powered characters about models and their visualizations.
- Explore model embeddings through interactive t-SNE views.
- Use a story to connect explanations of predictions and decision boundaries.
Links
Paper Link
Enhancing AI Explainability for Non-technical Users with LLM-Driven Narrative Gamification (ACM DL).
Video figure
Short visual overview of the system.
Video demo
Full walkthrough of the interaction design.
Skills
- Python
- t-SNE
- XAI Visualization
- Large Language Models (LLMs)
- Gamification
- AI Explainability
- OpenAI API
- Javascript
- D3.js
Keywords
- Human-Computer Interaction
- Information Visualization
- Explainable AI
- Gamification
- t-SNE
- Large Language Models
- XAI Visualization
Authors
Yuzhe You, Jian Zhao