EditCLEVR uses before/after scene pairs with one known object edit to show that object-centric models fail to decode exactly the intended single-object attribute change under a CoGenT color-shape shift, even with ground-truth masks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
EditCLEVR: A Paired-Scene Intervention Benchmark for Compositional Faithfulness of Object-Centric Representations
EditCLEVR uses before/after scene pairs with one known object edit to show that object-centric models fail to decode exactly the intended single-object attribute change under a CoGenT color-shape shift, even with ground-truth masks.