New GLARE interface uses LLMs to query global explanations of vision models

Researchers have developed GLARE, a new natural language interface designed to make global explanations of vision models more accessible. This system uses a large language model to translate user questions into SQL queries, which then access local explanation data. The interface provides responses in natural language, augmented with statistics and visualizations, to help users understand complex model behaviors across different datasets and contexts. Evaluations indicate that GLARE significantly enhances the usability of global explanations for human-centered explainable AI (XAI). AI

IMPACT Enhances accessibility and usability of global explanations for vision models, aiding human-centered XAI.

RANK_REASON The cluster contains a research paper detailing a new system for querying model explanations. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• LLM
• ScienceCast
• SQL
• XAI

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →