Evaluating Object-Centric Models beyond Object Discovery
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915784144453632 |
|---|---|
| author | Singh, Krishnakant Schaub-Meyer, Simone Roth, Stefan |
| author_facet | Singh, Krishnakant Schaub-Meyer, Simone Roth, Stefan |
| contents | Object-centric learning (OCL) aims to learn structured scene representations that support compositional generalization and robustness to out-of-distribution (OOD) data. However, OCL models are often not evaluated regarding these goals. Instead, most prior work focuses on evaluating OCL models solely through object discovery and simple reasoning tasks, such as probing the representation via image classification. We identify two limitations in existing benchmarks: (1) They provide limited insights on the representation usefulness of OCL models, and (2) localization and representation usefulness are assessed using disjoint metrics. To address (1), we use instruction-tuned VLMs as evaluators, enabling scalable benchmarking across diverse VQA datasets to measure how well VLMs leverage OCL representations for complex reasoning tasks. To address (2), we introduce a unified evaluation task and metric that jointly assess localization (where) and representation usefulness (what), thereby eliminating inconsistencies introduced by disjoint evaluation. Finally, we include a simple multi-feature reconstruction baseline as a reference point. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_07532 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Evaluating Object-Centric Models beyond Object Discovery Singh, Krishnakant Schaub-Meyer, Simone Roth, Stefan Computer Vision and Pattern Recognition Machine Learning Object-centric learning (OCL) aims to learn structured scene representations that support compositional generalization and robustness to out-of-distribution (OOD) data. However, OCL models are often not evaluated regarding these goals. Instead, most prior work focuses on evaluating OCL models solely through object discovery and simple reasoning tasks, such as probing the representation via image classification. We identify two limitations in existing benchmarks: (1) They provide limited insights on the representation usefulness of OCL models, and (2) localization and representation usefulness are assessed using disjoint metrics. To address (1), we use instruction-tuned VLMs as evaluators, enabling scalable benchmarking across diverse VQA datasets to measure how well VLMs leverage OCL representations for complex reasoning tasks. To address (2), we introduce a unified evaluation task and metric that jointly assess localization (where) and representation usefulness (what), thereby eliminating inconsistencies introduced by disjoint evaluation. Finally, we include a simple multi-feature reconstruction baseline as a reference point. |
| title | Evaluating Object-Centric Models beyond Object Discovery |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2602.07532 |