作者位次 #3/6ACL 2026Poster· Natural Language Processing
Believing without Seeing: Quality Scores for Contextualizing Vision-Language Model Explanations
Keyu He · Tejas Srinivasan · Brihi Joshi · Xiang Ren · Jesse Thomason · Swabha Swayamdipta
摘要
When people query Vision-Language Models (VLMs) but cannot see the accompanying visual context (e.g. for blind and low-vision users), augmenting VLM predictions with natural language explanations can signal which model predictions are reliable. However, prior work has found that explanations can...