AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

Deciphering the Role of Representation Disentanglement: Investigating Compositional Generalization in CLIP Models

Reza Abbasi, Mohammad Rohban, Mahdieh Soleymani Baghshah

ECCV 2024 · Label-efficient and robust learning › OOD and anomaly detection · approach: Robust fine-tuning for OOD generalization · 7 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
higher text and image representation disentanglement (more disentangled CLIP)CLIP models · ImageNet-AOC-OoD accuracyImproved
lower soft rank of attribute-object embeddingsCLIP models · ImageNet-AOC-OoD accuracyImproved
train on ClipScore-filtered CommonPool-XL instead of unfilteredCLIP (CommonPool-XL) · ImageNet-AOC-OoD accuracyImproved
larger training dataset sizeCLIP models · ImageNet-AOC-OoD accuracyMixed
SigLIP (WebLI) improved backbone/ID accuracySigLIP · ImageNet-AOC-OoD accuracyNo clear effect
higher C-OoD models switch fewer dimensions for color changeCLIP models · Shapes3Ddimensions switched: 250 (YFCC RN50) vs 40 (LAION ViT-L/14)Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy