Ordering word prediction by a dependency tree instead of left-to-right improves vision-language compositional understanding across five benchmarks.
head” word and its “dependent
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Causal Graphical Models for Vision-Language Compositional Understanding
Ordering word prediction by a dependency tree instead of left-to-right improves vision-language compositional understanding across five benchmarks.