Words or Vision: Do Vision-Language Models Have Blind Faith in Text?

Connections are not prepared for this record yet, or its source metadata has changed. The original record remains available while background snapshots are rebuilt.