Research map / Vision-language and multimodal / Human and embodied modeling
Visual-Semantic Matching by Exploring High-Order Attention and Distraction
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 |
|---|---|---|---|
| Add relation information to object-only scene graphs (ablation) | Visual-semantic matching model · MS-COCO 1K | sentence retrieval R@1: 70.9 -> 74.1 | Helped |
| Add relation information to object-only scene graphs (ablation) | Visual-semantic matching model · MS-COCO 1K | image retrieval R@1: 52.3 -> 61.8 | Helped |
| Add attribute information to graphs with objects and relations (ablation) | Visual-semantic matching model · MS-COCO 1K | sentence retrieval R@1: 74.1 -> 77.0 | Helped |
| Add attribute information to graphs with objects and relations (ablation) | Visual-semantic matching model · MS-COCO 1K | image retrieval R@1: 61.8 -> 65.1 | Helped |
| Remove cross-graph attention (ablation) | Visual-semantic matching model · MS-COCO 1K | sentence retrieval R@1: 77.0 -> 46.9 | Hurt |
| Remove cross-graph attention (ablation) | Visual-semantic matching model · MS-COCO 1K | image retrieval R@1: 65.1 -> 47.5 | Hurt |
| Remove node attention (ablation) | Visual-semantic matching model · MS-COCO 1K | sentence retrieval R@1: 77.0 -> 56.7 | Hurt |
| Remove node attention (ablation) | Visual-semantic matching model · MS-COCO 1K | image retrieval R@1: 65.1 -> 51.3 | Hurt |
| Add entropy-based distraction re-ranking to the top-10 results | Visual-semantic matching model · MS-COCO 1K | sentence retrieval R@1: 77.0 -> 77.8 | Helped |
| Add entropy-based distraction re-ranking to the top-10 results | Visual-semantic matching model · MS-COCO 5K | image retrieval R@1: 39.4 -> 40.5 | Helped |
| Add entropy-based distraction re-ranking to the top-10 results | Visual-semantic matching model · Flickr30K | sentence retrieval R@1: 70.8 -> 70.8 | No clear effect |
Closest papers by meaning
- Composing Object Relations and Attributes for Image-Text Matching (CVPR 2024 · similarity 0.89)
- ResMatch: Residual Attention Learning for Feature Matching (AAAI 2024 · similarity 0.86)
- SCENIR: Visual Semantic Clarity through Unsupervised Scene Graph Retrieval (ICML 2025 · similarity 0.86)
- Remodeling Semantic Relationships in Vision-Language Fine-Tuning (AAAI 2026 · similarity 0.86)
- Identification of Necessary Semantic Undertakers in the Causal View for Image-Text Matching (AAAI 2024 · similarity 0.86)
- Text Embedding is Not All You Need: Attention Control for Text-to-Image Semantic Alignment with Text Self-Attention Maps (CVPR 2025 · 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.