AnswerMap: Training-Free Black-Box Spatial Rationale Hits 0.85 AUC vs 0.38 for Attention

Mohamed Eltahir · hf · 2026-09-29

Researchers introduce AnswerMap, a training-free, task-agnostic, black-box visual rationale method for VLMs.

Method

Faithfulness validation (4 models × 3 query distributions)

Three read-out uses

Original post →

More from Multimodal

Multimodal channel →