Research map / Vision-language and multimodal / Cross-modal learning
Polysemous Visual-Semantic Embedding for Cross-Modal Retrieval
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 multiple embeddings instead of a single embedding (ablation) | PVSE · MS-COCO, TGIF, and MRW | rsum | Helped |
| Add the Local Feature Transformer to global-feature-only embeddings (K=0) (ablation) | PVSE · MS-COCO, TGIF, and MRW | rsum | Helped |
| Add global features to locally-guided features (ablation) | PVSE · MS-COCO and MRW | rsum | Helped |
| Combine global and locally-guided features with residual learning rather than concatenation (ablation) | PVSE · MS-COCO and MRW | rsum | Helped |
| Use the MIL objective rather than concatenating embeddings and using standard triplet ranking (ablation) | PVSE · MS-COCO | rsum: p=0.046 | Helped |
| Use the MIL objective rather than concatenating embeddings and using standard triplet ranking (ablation) | PVSE · MRW | rsum: p=0.015 | Helped |
| Include the diversity loss (ablation) | PVSE · MS-COCO | rsum | Helped |
| Include the MMD domain discrepancy loss (ablation) | PVSE · MS-COCO | rsum | Helped |
| Vary the relative weights of the diversity and MMD losses (ablation) | PVSE · MS-COCO | rsum | No clear effect |
| Use a more sophisticated multi-head self-attention module instead of the two-layer perceptron (ablation) | PIE-Net · Not specified | retrieval performance | No clear effect |
Closest papers by meaning
- Polysemic Semantic Instance Network for Cross-Modal Hashing (AAAI 2026 · similarity 0.90)
- Adaptive Prompt-Based Semantic Embedding with Inspire Potential of Implicit Knowledge for Cross-Modal Retrieval (AAAI 2025 · similarity 0.88)
- Same Concept, Different Directions: Cross-Modal Feature Heterogeneity in Sparse Autoencoders (NeurIPS 2026 · similarity 0.87)
- VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Tasks (ICLR 2025 · similarity 0.86)
- WikiDO: A New Benchmark Evaluating Cross-Modal Retrieval for Vision-Language Models (NeurIPS 2024 · similarity 0.86)
- Asymmetric Visual Semantic Embedding Framework for Efficient Vision-Language Alignment (AAAI 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.