Research map / Graph learning / Knowledge and temporal graphs
Unbiased Scene Graph Generation From Biased Training
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Apply TDE at inference | MOTIFS†-SUM · Visual Genome PredCls | mR@100: 15.8 -> 29.1 | Helped |
| Apply TDE at inference | MOTIFS†-SUM · Visual Genome SGCls | mR@100: 8.5 -> 14.9 | Helped |
| Apply TDE at inference | MOTIFS†-SUM · Visual Genome SGDet | mR@100: 6.8 -> 9.8 | Helped |
| Apply TDE at inference | VTransE†-SUM · Visual Genome SGDet | mR@100: 6.0 -> 10.5 | Helped |
| Apply TDE at inference | VCTree†-SUM · Visual Genome SGDet | mR@100: 6.9 -> 11.1 | Helped |
| Apply TDE at inference | MOTIFS†-SUM · Visual Genome zero-shot PredCls | R@50 / R@100: 10.9 / 14.5 -> 14.4 / 18.2 | Helped |
| Apply TDE at inference | MOTIFS†-SUM · Visual Genome zero-shot SGDet | R@50 / R@100: 0.1 / 0.2 -> 2.3 / 2.9 | Helped |
| Apply TDE at inference | MOTIFS†-SUM · Sentence-to-Graph Retrieval, gallery 1,000 | R@20: 11.6 -> 17.0 | Helped |
| Use focal loss for debiasing (ablation) | MOTIFS†-SUM · Visual Genome PredCls | mR@100: 15.8 -> 15.0 | Hurt |
| Reweight training examples by inverse predicate frequency (ablation) | MOTIFS†-SUM · Visual Genome zero-shot SGDet | R@50 / R@100: 0.1 / 0.2 -> 0.0 / 0.0 | Hurt |
| Resample training examples by inverse predicate frequency (ablation) | MOTIFS†-SUM · Visual Genome PredCls | mR@100: 15.8 -> 20.0 | Helped |
| Use only the object-feature-to-predicate branch at inference (ablation) | MOTIFS†-SUM · Visual Genome PredCls | mR@100: 15.8 -> 17.6 | Helped |
Closest papers by meaning
- Towards Unbiased and Robust Spatio-Temporal Scene Graph Generation and Anticipation (CVPR 2025 · similarity 0.88)
- Panoptic Scene Graph Generation with Semantics-Prototype Learning (AAAI 2024 · similarity 0.88)
- TD²-Net: Toward Denoising and Debiasing for Video Scene Graph Generation (AAAI 2024 · similarity 0.88)
- HiKER-SGG: Hierarchical Knowledge Enhanced Robust Scene Graph Generation (CVPR 2024 · similarity 0.88)
- Conformal Prediction and MLLM aided Uncertainty Quantification in Scene Graph Generation (CVPR 2025 · similarity 0.86)
- Adaptive Self-training Framework for Fine-grained Scene Graph Generation (ICLR 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.