The Language Interpretability Tool: Extensible, Interactive Visualizations and Analysis for NLP Models

2020/08/12 by Ian Tenney, James Wexler, Tenney, Ian +19 · 5 citations
Computer Science · #Adversarial Robustness in Machine Learning #Computation and Language (cs.CL) #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2008.05122

openalex publication_date 2020/08/12 · openalex created_date 2020/08/18 · openalex updated_date 2026/07/28

Abstract

We present the Language Interpretability Tool (LIT), an open-source platform for visualization and understanding of NLP models. We focus on core questions about model behavior: Why did my model make this prediction? When does it perform poorly? What happens under a controlled change in the input? LIT integrates local explanations, aggregate analysis, and counterfactual generation into a streamlined, browser-based interface to enable rapid exploration and error analysis. We include case studies for a diverse set of workflows, including exploring counterfactuals for sentiment analysis, measuring gender bias in coreference systems, and exploring local behavior in text generation. LIT supports a wide range of models--including classification, seq2seq, and structured prediction--and is highly extensible through a declarative, framework-agnostic API. LIT is under active development, with code and full documentation available at https://github.com/pair-code/lit.

Citations

• Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word\n Embeddings
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• InterpretML: A Unified Framework for Machine Learning Interpretability

Cited by

• VeriLLMed: Interactive Visual Debugging of Medical Large Language Models with Knowledge Graphs
• ClinStructor: AI-Powered Structuring of Unstructured Clinical Texts
• How AI Fails: An Interactive Pedagogical Tool for Demonstrating Dialectal Bias in Automated Toxicity Models
• Exploring the Landscape of Fairness Interventions in Software Engineering
• Critique of impure reason: Unveiling the reasoning behaviour of medical large language models

Related

• AllenNLP Interpret: A Framework for Explaining Predictions of NLP Models
• Visualizing and Explaining Language Models
• Multi-resolution Interpretation and Diagnostics Tool for Natural Language Classifiers
• On Behalf of the Stakeholders: Trends in NLP Model Interpretability in the Era of LLMs
• Visual Analytics for Fine-grained Text Classification Models and Datasets