Research map / Vision-language and multimodal / Cross-modal learning
Incomplete Multimodality-Diffused Emotion Recognition
Yuanzhi Wang, Yong Li, Zhen Cui
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 |
|---|---|---|---|
| add the diffusion score model s_k for modality recovery (ablation) | IMDer · CMU-MOSI random missing avg | ACC2: 75.8 -> 79.6 | Improved |
| add the reconstruction/refinement module D_k (ablation) | IMDer · CMU-MOSI random missing avg | ACC2: 77.8 -> 79.6 | Improved |
| add the diffusion score model s_k (ablation) | IMDer · CMU-MOSEI random missing avg | ACC2: 78.0 -> 80.9 | Improved |
| add the reconstruction module D_k (ablation) | IMDer · CMU-MOSEI random missing avg | ACC2: 79.8 -> 80.9 | Improved |
| add both s_k and D_k versus neither (ablation) | IMDer · CMU-MOSI random missing avg | ACC2: 74.5 -> 79.6 | Improved |
| score-model recovery versus reconstruction-only recovery (ablation) | IMDer · CMU-MOSEI random missing avg | ACC2: 78.0 (s_k removed) vs 79.8 (D_k removed) | Mixed |
| diffusion-based recovery versus graph completion (GCNet) | IMDer · CMU-MOSI random missing MR 0.7 | ACC2: 65.7 -> 71.9 | Improved |
| diffusion-based recovery versus GCNet when only language is available | IMDer · CMU-MOSI fixed {l} | ACC2: 83.7 -> 84.8 | Improved |
| IMDer versus GCNet with full modalities | IMDer · CMU-MOSEI MR 0.0 | ACC2: 85.2 -> 85.1 | No clear effect |
| condition recovery on available modalities (conditional vs unconditional) (ablation) | IMDer · CMU-MOSEI t-SNE | class separability (qualitative) | Improved |
Closest papers by meaning
- Federated Dialogue-Semantic Diffusion for Emotion Recognition under Incomplete Modalities (NeurIPS 2025 · similarity 0.90)
- Multi-to-Single: Reducing Multimodal Dependency in Emotion Recognition Through Contrastive Learning (AAAI 2025 · similarity 0.89)
- A Unified Self-Distillation Framework for Multimodal Sentiment Analysis with Uncertain Missing Modalities (AAAI 2024 · similarity 0.88)
- OV-MER: Towards Open-Vocabulary Multimodal Emotion Recognition (ICML 2025 · similarity 0.87)
- EMOE: Modality-Specific Enhanced Dynamic Emotion Experts (CVPR 2025 · similarity 0.87)
- Toward Robust Incomplete Multimodal Sentiment Analysis via Hierarchical Representation Learning (NeurIPS 2024 · 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.
