Research map / Vision-language and multimodal / Cross-modal learning
CMAD: Correlation-Aware and Modalities-Aware Distillation for Multimodal Sentiment Analysis with Missing Modalities
Yan Zhuang, Minhao Liu, Wei Bai, Yanru Zhang, Xiaoyue Zhang, Jiawen Deng, Fuji Ren
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 CAFD (correlation-aware feature distillation) (ablation) | CMAD · MOSEI | avg accuracy/F1: 76.2/75.5 -> 77.0/76.4 | Improved |
| add CAFD (ablation) | CMAD · IEMOCAP | avg accuracy/F1: 78.8/76.8 -> 79.2/77.8 | Improved |
| add MAR (modality-aware regularization weights) (ablation) | CMAD · MOSEI | avg accuracy/F1: 75.5/75.5 -> 77.0/76.4 | Improved |
| add MAR (ablation) | CMAD · IEMOCAP | avg accuracy/F1: 78.9/76.4 -> 79.2/77.8 | Improved |
| eliminate sample difficulty when computing modality weights (CMAD vs CMADt) | CMAD · MOSEI | avg F1: 75.2 -> 76.4 | Improved |
| eliminate sample difficulty (CMADw vs MMANet, both focus on hard combinations) | CMADw · MOSEI | avg F1: 73.2 -> 75.6 | Improved |
| use CMAD instead of second-best UMDF | CMAD · CHERMA | avg improvement: 1.4 | Improved |
| combine feature and correlation consistency instead of teacher-teacher to student-student transfer (ablation) | CAFD · MOSEI | avg accuracy/F1: 76.23/75.43 -> 77.03/76.39 | Improved |
| combine feature and correlation consistency instead of teacher-teacher to student-student transfer (ablation) | CAFD · IEMOCAP | avg accuracy/F1: 77.78/76.39 -> 79.23/77.84 | Improved |
| CMAD vs baselines on MUStARD and UR-FUNNY | CMAD · MUStARD | avg improvement: 1.9% | Improved |
| CMAD vs baselines on UR-FUNNY | CMAD · UR-FUNNY | avg improvement: 0.5% | Improved |
Later papers that compared with this method
1 later accepted paper reports CMA-D as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Path-level Hindsight Instructions for Semantic Exploration in Vision-Language Navigation (ECCV 2026 · better on 3 of 3 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- A Unified Self-Distillation Framework for Multimodal Sentiment Analysis with Uncertain Missing Modalities (AAAI 2024 · similarity 0.91)
- Correlation-Decoupled Knowledge Distillation for Multimodal Sentiment Analysis with Incomplete Modalities (CVPR 2024 · similarity 0.91)
- Toward Robust Incomplete Multimodal Sentiment Analysis via Hierarchical Representation Learning (NeurIPS 2024 · similarity 0.88)
- SemMSA: Latent Semantic-Aided Robust Multimodal Sentiment Analysis with Incomplete Data (NeurIPS 2026 · similarity 0.88)
- DLF: Disentangled-Language-Focused Multimodal Sentiment Analysis (AAAI 2025 · similarity 0.88)
- Towards Robust Multimodal Sentiment Analysis with Incomplete Data (NeurIPS 2024 · similarity 0.88)
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.
