LURE reduces object hallucination in LVLMs by 23% via post-hoc revision informed by co-occurrence, uncertainty, and text position analysis.
Currently, the CHAIR metric can only be applied to the COCO dataset, which limits its usability beyond that dataset
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2023 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Analyzing and Mitigating Object Hallucination in Large Vision-Language Models
LURE reduces object hallucination in LVLMs by 23% via post-hoc revision informed by co-occurrence, uncertainty, and text position analysis.