Research map / Vision-language and multimodal / Cross-modal learning
Dual Adversarial Graph Neural Networks for Multi-label Cross-modal Retrieval
Shengsheng Qian, Dizhan Xue, Huaiwen Zhang, Quan Fang, Changsheng Xu
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 |
|---|---|---|---|
| use DAGNN instead of DSCMR | cross-modal retrieval · NUS-WIDE | mAP, Average: 0.722 -> 0.757 | Improved |
| use DAGNN instead of DSCMR | cross-modal retrieval · MIRFlickr | mAP, Average: 0.775 -> 0.811 | Improved |
| add classification loss Lcla (ablation) | DAGNN · NUS-WIDE | mAP, Average: 0.520 -> 0.757 | Improved |
| add classification loss Lcla (ablation) | DAGNN · MIRFlickr | mAP, Average: 0.723 -> 0.811 | Improved |
| add modality invariance loss Lmdi (ablation) | DAGNN · NUS-WIDE | mAP, Average: 0.755 -> 0.757 | Improved |
| add modality invariance loss Lmdi (ablation) | DAGNN · MIRFlickr | mAP, Average: 0.807 -> 0.811 | Improved |
| add adversarial loss Ladv (ablation) | DAGNN · NUS-WIDE | mAP, Average: 0.750 -> 0.757 | Improved |
| add adversarial loss Ladv (ablation) | DAGNN · MIRFlickr | mAP, Average: 0.805 -> 0.811 | Improved |
| use cross-reconstruction instead of discriminating common image and text representations (ablation) | DAGNN · NUS-WIDE | mAP, Average: 0.747 -> 0.757 | Improved |
| use cross-reconstruction instead of discriminating common image and text representations (ablation) | DAGNN · MIRFlickr | mAP, Average: 0.802 -> 0.811 | Improved |
| use Multi-hop GNN instead of a linear classifier (ablation) | DAGNN · NUS-WIDE | mAP, Average: 0.734 -> 0.757 | Improved |
| use Multi-hop GNN instead of a linear classifier (ablation) | DAGNN · MIRFlickr | mAP, Average: 0.760 -> 0.811 | Improved |
Closest papers by meaning
- Integrating Multi-Label Contrastive Learning With Dual Adversarial Graph Neural Networks for Cross-Modal Retrieval (TPAMI 2022 · similarity 0.95)
- Joint Feature Synthesis and Embedding: Adversarial Cross-Modal Retrieval Revisited (TPAMI 2020 · similarity 0.91)
- Self-Supervised Adversarial Hashing Networks for Cross-Modal Retrieval (CVPR 2018 · similarity 0.88)
- Learning Cross-Aligned Latent Embeddings for Zero-Shot Cross-Modal Retrieval (AAAI 2020 · similarity 0.88)
- Multimodal Graph Neural Architecture Search under Distribution Shifts (AAAI 2024 · similarity 0.86)
- DiCA: Disambiguated Contrastive Alignment for Cross-Modal Retrieval with Partial Labels (AAAI 2025 · similarity 0.85)
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.
