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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| higher text and image representation disentanglement (more disentangled CLIP) | CLIP models · ImageNet-AO | C-OoD accuracy | Improved |
| lower soft rank of attribute-object embeddings | CLIP models · ImageNet-AO | C-OoD accuracy | Improved |
| train on ClipScore-filtered CommonPool-XL instead of unfiltered | CLIP (CommonPool-XL) · ImageNet-AO | C-OoD accuracy | Improved |
| larger training dataset size | CLIP models · ImageNet-AO | C-OoD accuracy | Mixed |
| SigLIP (WebLI) improved backbone/ID accuracy | SigLIP · ImageNet-AO | C-OoD accuracy | No clear effect |
| higher C-OoD models switch fewer dimensions for color change | CLIP models · Shapes3D | dimensions switched: 250 (YFCC RN50) vs 40 (LAION ViT-L/14) | Improved |
Closest papers by meaning
- Does CLIP’s generalization performance mainly stem from high train-test similarity? (ICLR 2024 · similarity 0.91)
- When and How Does CLIP Enable Domain and Compositional Generalization? (ICML 2025 · similarity 0.90)
- Does Data Scaling Lead to Visual Compositional Generalization? (ICML 2025 · similarity 0.88)
- Demystifying CLIP Data (ICLR 2024 · similarity 0.88)
- Interpreting CLIP with Sparse Linear Concept Embeddings (SpLiCE) (NeurIPS 2024 · similarity 0.87)
- CLIP in Mirror: Disentangling text from visual images through reflection (NeurIPS 2024 · similarity 0.87)
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.
