Research map / Graph learning / Knowledge and temporal graphs
Adaptive Fine-Grained Predicates Learning for Scene Graph Generation
Xinyu Lyu, Lianli Gao, Pengpeng Zeng, Heng Tao Shen, Jingkuan Song
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 |
|---|---|---|---|
| add Category Discriminating Loss (CDL) (ablation) | Transformer · VG-SGG PredCls | mR@100: 17.5 -> 35.4 | Improved |
| add Entity Discriminating Loss (EDL) on top of CDL (ablation) | Transformer · VG-SGG PredCls | mR@100: 35.4 -> 40.3 | Improved |
| add Entity Discriminating Loss (EDL) on top of CDL (ablation) | VCTree · VG-SGG PredCls | mR@100: 35.3 -> 40.2 | Improved |
| add Entity Discriminating Loss (EDL) on top of CDL (ablation) | Motif · VG-SGG PredCls | mR@100: 34.4 -> 37.5 | Improved |
| replace CDL with adaptive CDL-A (adaptive label softening) (ablation) | Transformer · VG-SGG PredCls | mR@100: 35.4 -> 38.9 | Improved |
| replace CDL with adaptive CDL-A (ablation) | VCTree · VG-SGG PredCls | mR@100: 35.3 -> 37.8 | Improved |
| replace CDL with adaptive CDL-A (ablation) | Motif · VG-SGG PredCls | mR@100: 34.4 -> 37.9 | Improved |
| replace CDL+EDL with CDL-A+EDL-A (full FGPL-A) (ablation) | VCTree · VG-SGG PredCls | mR@100: 40.2 -> 44.3 | Improved |
| replace CDL+EDL with CDL-A+EDL-A (full FGPL-A) (ablation) | Transformer · VG-SGG PredCls | mR@100: 40.3 -> 42.4 | Improved |
| replace CDL+EDL with CDL-A+EDL-A (full FGPL-A) (ablation) | Motif · VG-SGG SGDet | mR@100: 18.2 -> 19.8 | Improved |
| use FGPL-A instead of dataset-frequency re-weighting | Transformer · VG-SGG PredCls | mR@100: 34.4 -> 42.4 | Improved |
| use FGPL-A instead of re-weighting | VCTree · VG-SGG PredCls | DP@10: 36.5 -> 39.3 | Improved |
Closest papers by meaning
- Noise-Guided Predicate Representation Extraction and Diffusion-Enhanced Discretization for Scene Graph Generation (ICML 2025 · similarity 0.90)
- RA-SGG: Retrieval-Augmented Scene Graph Generation Framework via Multi-Prototype Learning (AAAI 2025 · similarity 0.89)
- Leveraging Predicate and Triplet Learning for Scene Graph Generation (CVPR 2024 · similarity 0.89)
- Adaptive Self-training Framework for Fine-grained Scene Graph Generation (ICLR 2024 · similarity 0.89)
- LLM4SGG: Large Language Models for Weakly Supervised Scene Graph Generation (CVPR 2024 · similarity 0.86)
- Multi-Prototype Space Learning for Commonsense-Based Scene Graph Generation (AAAI 2024 · similarity 0.86)
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.
