Research map / Vision-language and multimodal / CLIP and zero-shot
Learning Cross-Aligned Latent Embeddings for Zero-Shot Cross-Modal Retrieval
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 cross-reconstruction loss (ablation) | LCALE · Wikipedia image-text | Avg MAP: 0.250 -> 0.362 | Helped |
| add Wasserstein cross-alignment loss (ablation) | LCALE · Wikipedia image-text | Avg MAP: 0.333 -> 0.362 | Helped |
| add MMD cross-alignment loss (ablation) | LCALE · Wikipedia image-text | Avg MAP: 0.347 -> 0.362 | Helped |
| add cycle-consistency loss (ablation) | LCALE · Wikipedia image-text | Avg MAP: 0.339 -> 0.362 | Helped |
| add cross-reconstruction loss (ablation) | LCALE · Sketchy | MAP: 0.422 -> 0.476 | Helped |
| add Wasserstein cross-alignment loss (ablation) | LCALE · Sketchy | MAP: 0.454 -> 0.476 | Helped |
| add MMD cross-alignment loss (ablation) | LCALE · Sketchy | MAP: 0.395 -> 0.476 | Helped |
| add cycle-consistency loss (ablation) | LCALE · Sketchy | MAP: 0.458 -> 0.476 | Helped |
| replace GAN-based class-embedding space (DADN) with VAE latent embeddings | ZS-CMR · Wikipedia image-text | Avg MAP: 0.298 -> 0.362 | Helped |
| replace SEM-PCYC with LCALE | ZS image-sketch · Sketchy | MAP: 0.349 -> 0.476 | Helped |
| vary latent embedding dimension (25-250); peak near 60-100 (ablation) | LCALE · Wikipedia image-text | MAP | Mixed |
| binary quantized embeddings instead of real-valued (ablation) | LCALE · Sketchy | MAP: 0.476 -> 0.397 | Hurt |
Closest papers by meaning
- Same Concept, Different Directions: Cross-Modal Feature Heterogeneity in Sparse Autoencoders (NeurIPS 2026 · similarity 0.87)
- Language-only Training of Zero-shot Composed Image Retrieval (CVPR 2024 · similarity 0.87)
- With Limited Data for Multimodal Alignment, Let the STRUCTURE Guide You (NeurIPS 2025 · similarity 0.87)
- CL2CM: Improving Cross-Lingual Cross-Modal Retrieval via Cross-Lingual Knowledge Transfer (AAAI 2024 · similarity 0.87)
- WikiDO: A New Benchmark Evaluating Cross-Modal Retrieval for Vision-Language Models (NeurIPS 2024 · similarity 0.87)
- Adaptive Prompt-Based Semantic Embedding with Inspire Potential of Implicit Knowledge for Cross-Modal Retrieval (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.