Research map / Vision-language and multimodal / Cross-modal learning
Learning Semantic Relationship Among Instances for Image-Text Matching
Read the paperCheck 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 instance-level cross-modal and intra-modal relation modeling (ablation) | HREM image-text matching · Flickr30K | R@1 IMG → TEXT: 81.6 -> 83.3 | Helped |
| Add the connection matrix A (ablation) | HREM image-text matching · Flickr30K | R@1 IMG → TEXT: 81.4 -> 83.3 | Helped |
| Add the relevance matrix S (ablation) | HREM image-text matching · Flickr30K | R@1 IMG → TEXT: 81.7 -> 83.3 | Helped |
| Add relevance regularization Lreg (ablation) | HREM image-text matching · Flickr30K | R@1 IMG → TEXT: 82.8 -> 83.3 | Helped |
| Add neighbor batch sampling (ablation) | HREM image-text matching · Flickr30K | R@1 IMG → TEXT: 82.8 -> 83.3 | Helped |
| Use the fusion mechanism instead of the full ensemble (ablation) | HREM · Flickr30K 1K | rSum: 520.4 -> 524.2 | Hurt |
| Add sample relation modeling during training | HREM image-text matching · Flickr30K hard samples (5%) | R@1 IMG → TEXT: 74.6 -> 76.0 | Helped |
| Add sample relation modeling during training | HREM image-text matching · Flickr30K infrequent samples (5%) | R@1 IMG → TEXT: 79.2 -> 84.0 | Helped |
| Add hierarchical instance-level relation modeling | HREM image-text matching · Flickr30K 1K | rSum: 513.5 -> 524.2 | Helped |
Closest papers by meaning
- Composing Object Relations and Attributes for Image-Text Matching (CVPR 2024 · similarity 0.89)
- Identification of Necessary Semantic Undertakers in the Causal View for Image-Text Matching (AAAI 2024 · similarity 0.87)
- Expanding the Scope of Negatives: Boosting Image-Text Matching with Negatives Distribution Guided Learning (AAAI 2025 · similarity 0.86)
- Text as Partial Constraint: Core–Residual Alignment for Robust Vision–Language Learning (NeurIPS 2026 · similarity 0.86)
- DH-Set: Improving Vision-Language Alignment with Diverse and Hybrid Set-Embeddings Learning (CVPR 2025 · 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.