Research map / Vision-language and multimodal / Cross-modal learning
Look, Imagine and Match: Improving Textual-Visual Cross-Modal Retrieval With Generative Models
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 image-to-text generation trained with XE loss (ablation) | GXN · MSCOCO 1K-image test, Image-to-Text | R@1: 64.0 -> 68.2 | Helped |
| Add image-to-text generation trained with XE loss (ablation) | GXN · MSCOCO 1K-image test, Text-to-Image | R@1: 53.6 -> 54.5 | Helped |
| Train image-to-text generation with mixed XE and RL losses (ablation) | GXN with image-to-text generation · MSCOCO 1K-image test, Image-to-Text | R@1: 68.2 -> 68.4 | Helped |
| Train image-to-text generation with mixed XE and RL losses (ablation) | GXN with image-to-text generation · MSCOCO 1K-image test, Text-to-Image | R@1: 54.5 -> 55.6 | Helped |
| Add text-to-image generative adversarial training (ablation) | GXN · MSCOCO 1K-image test, Text-to-Image | R@1: 53.6 -> 56.5 | Helped |
| Add text-to-image generative adversarial training (ablation) | GXN · MSCOCO 1K-image test, Image-to-Text | R@1: 64.0 -> 67.1 | Helped |
| Use a bidirectional GRU instead of a one-layer GRU sentence encoder (ablation) | GRU (VGG19) · MSCOCO 1K-image test, Image-to-Text | R@1: 51.4 -> 53.6 | Helped |
| Use a bidirectional GRU instead of a one-layer GRU sentence encoder (ablation) | GRU (VGG19) · MSCOCO 1K-image test, Text-to-Image | R@1: 39.1 -> 40.0 | Helped |
| Replace VGG19 with ResNet152 as the image encoder (ablation) | GRUBi · MSCOCO 1K-image test, Image-to-Text | R@1: 53.6 -> 64.0 | Helped |
| Replace VGG19 with ResNet152 as the image encoder (ablation) | GRUBi · MSCOCO 1K-image test, Text-to-Image | R@1: 40.0 -> 53.6 | Helped |
| Combine image-to-text and text-to-image generative training paths (ablation) | GXN · MSCOCO 1K-image test, Image-to-Text | R@1 | Helped |
| Combine image-to-text and text-to-image generative training paths (ablation) | GXN · MSCOCO 1K-image test, Text-to-Image | R@1 | Helped |
Closest papers by meaning
- WikiDO: A New Benchmark Evaluating Cross-Modal Retrieval for Vision-Language Models (NeurIPS 2024 · similarity 0.89)
- TIGeR: Unifying Text-to-Image Generation and Retrieval with Large Multimodal Models (ICLR 2025 · similarity 0.88)
- Recurrence-Enhanced Vision-and-Language Transformers for Robust Multimodal Document Retrieval (CVPR 2025 · similarity 0.88)
- Adaptive Prompt-Based Semantic Embedding with Inspire Potential of Implicit Knowledge for Cross-Modal Retrieval (AAAI 2025 · similarity 0.88)
- Cross the Gap: Exposing the Intra-modal Misalignment in CLIP via Modality Inversion (ICLR 2025 · similarity 0.88)
- Same Concept, Different Directions: Cross-Modal Feature Heterogeneity in Sparse Autoencoders (NeurIPS 2026 · similarity 0.88)
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.