Research map / Vision-language and multimodal / Cross-modal learning
OTKGE: Multi-modal Knowledge Graph Embeddings via Optimal Transport
Zongsheng Cao, Qianqian Xu, Zhiyong Yang, Yuan He, Xiaochun Cao, Qingming Huang
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 optimal transport fusion instead of mean (ablation) | OTKGE2 · FB-IMG | MRR: .828 -> .842 | Improved |
| use optimal transport fusion instead of concat (ablation) | OTKGE2 · FB-IMG | MRR: .838 -> .842 | Improved |
| use optimal transport fusion instead of mean (ablation) | OTKGE2 · WN9-IMG | MRR: .913 -> .923 | Improved |
| use optimal transport fusion instead of concat (ablation) | OTKGE2 · WN9-IMG | MRR: .916 -> .923 | Improved |
| add linguistic embeddings to visual-only (ablation) | OTKGE2 · FB-IMG | MRR: .835 -> .842 | Improved |
| add visual embeddings to linguistic-only (ablation) | OTKGE2 · FB-IMG | MRR: .836 -> .842 | Improved |
| add visual embeddings to linguistic-only (ablation) | OTKGE2 · WN9-IMG | MRR: .916 -> .923 | Improved |
| use multi-modal OTKGE instead of uni-modal TransE | OTKGE2 · FB-IMG | MRR: .712 -> .842 | Improved |
| use OTKGE instead of strongest multi-modal baseline MMKRL | OTKGE2 · FB-IMG | MRR: .827 -> .842 | Improved |
| use 2-Wasserstein instead of 1-Wasserstein distance | OTKGE2 · FB-IMG | MRR: .843 -> .842 | No clear effect |
Closest papers by meaning
- You CAN Teach an Old Dog New Tricks! On Training Knowledge Graph Embeddings (ICLR 2020 · similarity 0.88)
- LAFA: Multimodal Knowledge Graph Completion with Link Aware Fusion and Aggregation (AAAI 2024 · similarity 0.88)
- Non-Parametric Estimation of Multiple Embeddings for Link Prediction on Dynamic Knowledge Graphs (AAAI 2017 · similarity 0.88)
- Multiple Heads are Better than One: Mixture of Modality Knowledge Experts for Entity Representation Learning (ICLR 2025 · similarity 0.88)
- Mixed-Curvature Multi-Modal Knowledge Graph Completion (AAAI 2025 · similarity 0.87)
- Tokenization, Fusion, and Augmentation: Towards Fine-grained Multi-modal Entity Representation (AAAI 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.
