Research map / Vision-language and multimodal / Cross-modal learning
Deep Robust Unsupervised Multi-Modal Network
Yang Yang, Yi-Feng Wu, De-Chuan Zhan, Zhi-Bin Liu, Yuan Jiang
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 |
|---|---|---|---|
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · FLICKR25K I→T | MAP@50: .6416 -> .6266 | Hurt |
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · IAPR TC-12 I→T | MAP@50: .4323 -> .4605 | Improved |
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · WIKI I→T | MAP@50: .1567 -> .1781 | Improved |
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · NUS-WIDE I→T | MAP@50: .5009 -> .5626 | Improved |
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · FLICKR25K T→I | MAP@50: .5097 -> .6589 | Improved |
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · IAPR TC-12 T→I | MAP@50: .4578 -> .4974 | Improved |
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · WIKI T→I | MAP@50: .1722 -> .1887 | Improved |
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · NUS-WIDE T→I | MAP@50: .4776 -> .5548 | Improved |
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · WKG Game-Hub I→T L>1 | MAP@50: .2729 -> .3594 | Improved |
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · WKG Game-Hub I→T L>3 | MAP@50: .1829 -> .2629 | Improved |
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · WKG Game-Hub T→I L>1 | MAP@50: .2574 -> .3448 | Improved |
| replace threshold-based DRUMN-Thres with adaptive weight estimation (full DRUMN) (ablation) | DRUMN · WKG Game-Hub T→I L>3 | MAP@50: .1577 -> .2495 | Improved |
Closest papers by meaning
- Uncertainty-Aware Multi-modal Learning via Cross-Modal Random Network Prediction (ECCV 2022 · similarity 0.87)
- What Makes Training Multi-Modal Classification Networks Hard? (CVPR 2020 · similarity 0.86)
- Quantifying and Enhancing Multi-modal Robustness with Modality Preference (ICLR 2024 · similarity 0.85)
- Beyond Modality Fusion: Deep Ensembles for Multimodal Classification (NeurIPS 2026 · similarity 0.84)
- Multi-Modal Object Re-identification via Sparse Mixture-of-Experts (ICML 2025 · similarity 0.83)
- RedCore: Relative Advantage Aware Cross-Modal Representation Learning for Missing Modalities with Imbalanced Missing Rates (AAAI 2024 · similarity 0.83)
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.
