AnswerMap: Faithful Spatial Interpretability of VLMs from Answer Posteriors
AnswerMap builds black-box spatial rationales for vision-language models from output answer posteriors.
1 stories, newest first · one card per story
AnswerMap builds black-box spatial rationales for vision-language models from output answer posteriors.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views