Research map / Graph learning / Knowledge and temporal graphs
Factor Graph Attention
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 |
|---|---|---|---|
| Use factor graph attention instead of the strongest listed baseline | VisDial v0.9 · VisDial v0.9 | MRR: 0.6398 -> 0.6525 | Helped |
| Use FGA instead of SF-QIH-se-2 | Question generation · Question generation | MRR: 0.4060 -> 0.4138 | Helped |
| Use FGA instead of CorefNMN (ResNet-152) | VisDial v1.0 test-std · VisDial v1.0 test-std | MRR: 0.615 -> 0.637 | Helped |
| Replace VGG image features with F-RCNNx101 features | FGA · VisDial v0.9 | MRR: 0.6525 vs 0.6712 | Helped |
| Replace VGG image features with F-RCNNx101 features | 5×FGA · VisDial v1.0 test-std | MRR: 0.673 vs 0.693 | Helped |
| Add FGA attention (ablation) | VisDial v0.9 · VisDial v0.9 | MRR: 0.6249 -> 0.6525 | Helped |
| Add batch normalization over pairwise interactions (ablation) | FGA · VisDial v0.9 | MRR: 0.6301 -> 0.6525 | Helped |
| Add local entity interactions (ablation) | FGA · VisDial v0.9 | MRR: 0.6369 -> 0.6525 | Helped |
| Add local entity information (ablation) | FGA · VisDial v0.9 | MRR: 0.6425 -> 0.6525 | Helped |
| Add attention priors (ablation) | FGA · VisDial v0.9 | MRR: 0.6451 -> 0.6525 | Helped |
| Add the answer utility (ablation) | FGA · VisDial v0.9 | MRR: 0.6294 -> 0.6525 | Helped |
| Add history attention (ablation) | FGA · VisDial v0.9 | MRR: 0.6449 -> 0.6525 | Helped |
Closest papers by meaning
- Visual Dialog (CVPR 2017 · similarity 0.91)
- Iterative Context-Aware Graph Inference for Visual Dialog (CVPR 2020 · similarity 0.89)
- Reasoning Visual Dialogs With Structural and Partial Observations (CVPR 2019 · similarity 0.89)
- Visual Reference Resolution using Attention Memory for Visual Dialog (NeurIPS 2017 · similarity 0.89)
- Structure-Aware Multimodal Sequential Learning for Visual Dialog (AAAI 2024 · similarity 0.88)
- Visual Graph Arena: Evaluating Visual Conceptualization of Vision and Multimodal Large Language Models (ICML 2025 · 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.