Research map / Vision-language and multimodal / Cross-modal learning
Adaptive Cross-Modal Prototypes for Cross-Domain Visual-Language Retrieval
Yang Liu, Qingchao Chen, Samuel Albanie
Check 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 |
|---|---|---|---|
| Add adaptive cross-modal prototypes and mutual-information alignment to source-only retrieval | CE [37] · Open Narratives→COCO Narratives | t2v R1: 19.6 -> 22.3 | Improved |
| Add adaptive cross-modal prototypes and mutual-information alignment to source-only retrieval | CE [37] · COCO→COCO Narratives | v2t R1: 26.0 -> 28.0 | Improved |
| Add adaptive cross-modal prototypes and mutual-information alignment to source-only retrieval | CE [37] · MSRVTT→MSVD | t2v R1: 14.2 -> 16.6 | Improved |
| Add source prototype loss Ls (ablation) | ACP · MSRVTT→MSVD | t2v R1: 14.2 -> 15.7 | Improved |
| Add target prototype loss Lt (ablation) | ACP · MSRVTT→MSVD | t2v R1: 14.2 -> 15.4 | Improved |
| Add source prototype loss Ls (ablation) | ACP · MSRVTT→MSVD | v2t R1: 16.6 -> 20.4 | Improved |
| Add target prototype loss Lt (ablation) | ACP · MSRVTT→MSVD | v2t R1: 16.6 -> 19.8 | Improved |
| Add mutual-information loss LMI to source and target prototype losses (ablation) | ACP · MSRVTT→MSVD | t2v R1: 15.7 -> 16.6 | Improved |
| Add mutual-information loss LMI to source and target prototype losses (ablation) | ACP · MSRVTT→MSVD | v2t R1: 20.4 -> 22.1 | Improved |
Closest papers by meaning
- Retrieval Across Any Domains via Large-scale Pre-trained Model (ICML 2024 · similarity 0.90)
- WikiDO: A New Benchmark Evaluating Cross-Modal Retrieval for Vision-Language Models (NeurIPS 2024 · similarity 0.90)
- Dynamic Adapter with Semantics Disentangling for Cross-lingual Cross-modal Retrieval (AAAI 2025 · similarity 0.90)
- Adaptive Prompt-Based Semantic Embedding with Inspire Potential of Implicit Knowledge for Cross-Modal Retrieval (AAAI 2025 · similarity 0.89)
- CL2CM: Improving Cross-Lingual Cross-Modal Retrieval via Cross-Lingual Knowledge Transfer (AAAI 2024 · similarity 0.89)
- Cross the Gap: Exposing the Intra-modal Misalignment in CLIP via Modality Inversion (ICLR 2025 · similarity 0.89)
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.
