CIR benchmarks contain many unimodal shortcuts and noisy queries, leading to overestimation of models' multimodal composition capabilities.
When and why vision-language models behave like bags-of-words, and what to do about it? InICLR
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Do Composed Image Retrieval Benchmarks Require Multimodal Composition?
CIR benchmarks contain many unimodal shortcuts and noisy queries, leading to overestimation of models' multimodal composition capabilities.