Research map / Label-efficient and robust learning / Recognition and long-tail
Capturing Dependencies among Labels and Features for Multiple Emotion Tagging of Multimedia Data
Shan Wu, Shangfei Wang, Qiang Ji
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 |
|---|---|---|---|
| share features across labels with multi-task RBM instead of single-task RBM (ablation) | RBM · Music | accuracy: 0.546 -> 0.577 | Improved |
| share features across labels with multi-task RBM instead of single-task RBM (ablation) | RBM · NVIE | accuracy: 0.457 -> 0.495 | Improved |
| share features across labels with multi-task RBM instead of single-task RBM (ablation) | RBM · FilmStim | accuracy: 0.353 -> 0.394 | Improved |
| add high-order label dependency layer (method 1) to multi-task RBM (ablation) | multi-task RBM · NVIE | accuracy: 0.495 -> 0.502 | Improved |
| add high-order label dependency layer (method 1) to multi-task RBM (ablation) | multi-task RBM · FilmStim | accuracy: 0.394 -> 0.422 | Improved |
| add label dependency layer (method 1) to multi-task RBM (ablation) | multi-task RBM · Music | accuracy: 0.577 -> 0.584 | Improved |
| train all layers jointly in four-layer RBM (method 2) instead of separately (method 1) | proposed model · NVIE | accuracy: 0.502 -> 0.562 | Improved |
| train all layers jointly in four-layer RBM (method 2) instead of separately (method 1) | proposed model · FilmStim | accuracy: 0.422 -> 0.437 | Improved |
| train all layers jointly in four-layer RBM (method 2) instead of separately (method 1) | proposed model · Music | accuracy: 0.584 -> 0.585 | Improved |
| use single-task RBM (BR) instead of SVM(BR) | per-label classifier · Music | accuracy: 0.514 -> 0.546 | Improved |
| proposed method 2 vs Andreas et al. | emotion tagging · FilmStim | accuracy: 0.440 -> 0.437 | No clear effect |
Closest papers by meaning
- CARAT: Contrastive Feature Reconstruction and Aggregation for Multi-Modal Multi-Label Emotion Recognition (AAAI 2024 · similarity 0.85)
- AffectGPT: A New Dataset, Model, and Benchmark for Emotion Understanding with Multimodal Large Language Models (ICML 2025 · similarity 0.85)
- EmoGrowth: Incremental Multi-label Emotion Decoding with Augmented Emotional Relation Graph (ICML 2025 · similarity 0.85)
- Uncertain Multimodal Intention and Emotion Understanding in the Wild (CVPR 2025 · similarity 0.84)
- A Unified and Interpretable Emotion Representation and Expression Generation (CVPR 2024 · similarity 0.84)
- Beyond Single Emotion: Multi-label Approach to Conversational Emotion Recognition (AAAI 2025 · similarity 0.84)
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.
