SPARC selectively and progressively reinforces attention to relevant image tokens during decoding, improving both precision and recall in detailed image captioning compared to baselines and prior hallucination-mitigation methods.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Visual Attention Never Fades: Selective Progressive Attention ReCalibration for Detailed Image Captioning in Multimodal Large Language Models
SPARC selectively and progressively reinforces attention to relevant image tokens during decoding, improving both precision and recall in detailed image captioning compared to baselines and prior hallucination-mitigation methods.