Interactive Model Cards: A Human-Centered Approach to Model Documentation

Abstract

Deep learning models for natural language processing (NLP) are increasingly adopted and deployed by analysts without formal training in NLP or machine learning (ML). However, the documentation intended to convey the model's details and appropriate use is tailored primarily to individuals with ML or NLP expertise. To address this gap, we conduct a design inquiry into interactive model cards, which augment traditionally static model cards with affordances for exploring model documentation and interacting with the models themselves. Our investigation consists of an initial conceptual study with experts in ML, NLP, and AI Ethics, followed by a separate evaluative study with non-expert analysts who use ML models in their work. Using a semi-structured interview format coupled with a think-aloud protocol, we collected feedback from a total of 30 participants who engaged with different versions of standard and interactive model cards. Through a thematic analysis of the collected data, we identified several conceptual dimensions that summarize the strengths and limitations of standard and interactive model cards, including: stakeholders; design; guidance; understandability & interpretability; sensemaking & skepticism; and trust & safety. Our findings demonstrate the importance of carefully considered design and interactivity for orienting and supporting non-expert analysts using deep learning models, along with a need for consideration of broader sociotechnical contexts and organizational dynamics. We have also identified design elements, such as language, visual cues, and warnings, among others, that support interactivity and make non-interactive content accessible. We summarize our findings as design guidelines and discuss their implications for a human-centered approach towards AI/ML documentation.

arXiv commentTo appear at ACM FAccT'22

References in corpus (15)

• Array Programming with NumPy
• Model Cards for Model Reporting
• Explaining Explanations in AI
• Improving fairness in machine learning systems: What do industry practitioners need?
• Effect of Confidence and Explanation on Accuracy and Trust Calibration in AI-Assisted Decision Making
• Towards a Critical Race Methodology in Algorithmic Fairness
• Where Responsible AI meets Reality: Practitioner Perspectives on Enablers for shifting Organizational Practices
• Ethics-Based Auditing to Develop Trustworthy AI
• Hate Speech Detection and Racial Bias Mitigation in Social Media based on BERT model
• Explainability Fact Sheets: A Framework for Systematic Assessment of Explainable Approaches
• Human Factors in Model Interpretability: Industry Practices, Challenges, and Needs
• Everyday algorithm auditing: Understanding the power of everyday users in surfacing harmful algorithmic behaviors
• Racial categories in machine learning
• Why Does My Model Fail? Contrastive Local Explanations for Retail Forecasting
• Fits and Starts: Enterprise Use of AutoML and the Role of Humans in the Loop

Cited by in corpus (16)

• Auditing large language models: a three-layered approach
• Opening up ChatGPT: Tracking openness, transparency, and accountability in instruction-tuned text generators
• Understanding accountability in algorithmic supply chains
• Zeno: An Interactive Framework for Behavioral Evaluation of Machine Learning
• Certification Labels for Trustworthy AI: Insights From an Empirical Mixed-Method Study
• A Scoping Study of Evaluation Practices for Responsible AI Tools: Steps Towards Effectiveness Evaluations
• Aspirations and Practice of Model Documentation: Moving the Needle with Nudging and Traceability
• Design, Development, and Deployment of Context-Adaptive AI Systems for Enhanced End-User Adoption
• Making the Right Thing: Bridging HCI and Responsible AI in Early-Stage AI Concept Selection
• Help or Hinder? Evaluating the Impact of Fairness Metrics and Algorithms in Visualizations for Consensus Ranking
• Model Cards Revisited: Bridging the Gap Between Theory and Practice for Ethical AI Requirements
• DREAMS: A python framework for Training Deep Learning Models on EEG Data with Model Card Reporting for Medical Applications
• Talking About the Assumption in the Room
• MRM3: Machine Readable ML Model Metadata
• Transparency in Healthcare AI: Testing European Regulatory Provisions against Users' Transparency Needs
• Evaluation of AI Ethics Tools in Language Models: A Developers' Perspective Case Study