Research map / Detection and segmentation / Open-set and few-shot detection
Beyond Object Recognition: A New Benchmark towards Object Concept Learning
Yong-Lu Li, Yue Xu, Xinyu Xu, Xiaohan Mao, Yuan Yao, Siqi Liu, Cewu Lu
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 |
|---|---|---|---|
| Use attribute as intermediate input for affordance (alpha->beta structure) | DM baseline · OCL affordance recognition | mAP: 51.8 -> 52.6 | Improved |
| Use affordance as intermediate input for attribute (beta->alpha structure) | DM baseline · OCL affordance recognition | mAP: 51.8 -> 52.0 / 51.8 | No clear effect |
| Add ITE hinge loss with causal supervision | OCRN · OCL ITE reasoning | S_ITE: 9.5 -> 20.3 | Improved |
| Add ITE hinge loss with causal supervision | OCRN · OCL recognition | attribute mAP: 31.6 -> 31.5 | No clear effect |
| Remove deconfounding in OCRN (ablation) | OCRN · OCL validation | alpha, beta, ITE | Improved |
| Add instance-level losses (L_alpha, L_beta) (ablation) | OCRN · OCL validation | mAP: over 20 mAP | Improved |
| Add category-level losses (L_A, L_B) (ablation) | OCRN · OCL validation | mAP | Improved |
| Change feature dimension away from 1024 (ablation) | OCRN · OCL validation | mAP | Hurt |
| Use mean/max aggregation or random counterfactual feature instead of default (ablation) | OCRN · OCL validation | ITE | Hurt |
| Use hierarchical mapping with category-level supervision (HMa) instead of direct mapping | DM-V · OCL attribute recognition | mAP: 29.9 -> 28.6 | No clear effect |
Closest papers by meaning
- ConceptFactory: Facilitate 3D Object Knowledge Annotation with Object Conceptualization (NeurIPS 2024 · similarity 0.87)
- Rex-Thinker: Grounded Object Referring via Chain-of-Thought Reasoning (ICLR 2026 · similarity 0.86)
- Object-Centric Concept-Bottlenecks (NeurIPS 2025 · similarity 0.85)
- One-Shot Open Affordance Learning with Foundation Models (CVPR 2024 · similarity 0.85)
- OCRT: Boosting Foundation Models in the Open World with Object-Concept-Relation Triad (CVPR 2025 · similarity 0.85)
- Beyond the Doors of Perception: Vision Transformers Represent Relations Between Objects (NeurIPS 2024 · similarity 0.85)
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.
