CompART adds a composition loss on decomposed captions to regularize attention sums and improves multi-object grounding plus VQA across four VLM types and six benchmarks.
Scaling up visual and vision- language representation learning with noisy text su- pervision
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
The ART of Composition: Attention-Regularized Training for Compositional Visual Grounding
CompART adds a composition loss on decomposed captions to regularize attention sums and improves multi-object grounding plus VQA across four VLM types and six benchmarks.