Research map / Vision-language and multimodal / Cross-modal learning
Co-Regularized PLSA for Multi-Modal Learning
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 |
|---|---|---|---|
| Replace Link PLSA (identical topic compositions) with sKL coPLSA co-regularizer (ablation) | coPLSA · Wikipedia i2t, topic space | MAP: 0.247 -> 0.413 | Helped |
| Replace Link PLSA with sKL coPLSA (ablation) | coPLSA · TVGraz i2t, topic space | MAP: 0.349 -> 0.481 | Helped |
| Replace Link PLSA with sKL coPLSA (ablation) | coPLSA · TVGraz i2t, semantic space | MAP: 0.803 -> 0.850 | Helped |
| Replace Link PLSA with sKL coPLSA (ablation) | coPLSA · Wikipedia t2i, semantic space | MAP: 0.605 -> 0.724 | Helped |
| Replace SCM with sKL coPLSA (ablation) | coPLSA · Wikipedia i2t, topic space | MAP: 0.267 -> 0.413 | Helped |
| Replace SCM with sKL coPLSA (ablation) | coPLSA · TVGraz t2i, semantic space | MAP: 0.649 -> 0.850 | Helped |
| Use sKL instead of l1 co-regularizer (ablation) | coPLSA · Wikipedia i2t, topic space | MAP: 0.317 -> 0.413 | Helped |
| Use sKL instead of l1 co-regularizer (ablation) | coPLSA · TVGraz i2t, topic space | MAP: 0.359 -> 0.481 | Helped |
| Use sKL instead of l2 co-regularizer (ablation) | coPLSA · TVGraz i2t, topic space | MAP: 0.450 -> 0.481 | Helped |
| Use sKL instead of l2 co-regularizer (ablation) | coPLSA · Wikipedia i2t, semantic space | MAP: 0.706 -> 0.726 | Helped |
| Use l2 instead of l1 co-regularizer (ablation) | coPLSA · TVGraz i2t, topic space | MAP: 0.359 -> 0.450 | Helped |
| Add semantic abstraction (logistic regression mapping) to topic compositions (ablation) | sKL coPLSA · TVGraz i2t | MAP: 0.481 -> 0.850 | Helped |
Closest papers by meaning
- Removing Bias in Multi-modal Classifiers: Regularization by Maximizing Functional Entropies (NeurIPS 2020 · similarity 0.83)
- Probabilistic Embeddings for Cross-Modal Retrieval (CVPR 2021 · similarity 0.83)
- Cross-Modal Alignment via Variational Copula Modelling (ICML 2025 · similarity 0.82)
- Cross-Modal and Uni-Modal Soft-Label Alignment for Image-Text Retrieval (AAAI 2024 · similarity 0.82)
- SemMSA: Latent Semantic-Aided Robust Multimodal Sentiment Analysis with Incomplete Data (NeurIPS 2026 · similarity 0.81)
- RoDA: Robust Domain Alignment for Cross-Domain Retrieval Against Label Noise (AAAI 2025 · similarity 0.81)
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.