Research map / Vision-language and multimodal / Cross-modal learning
Tri-Subspaces Disentanglement for Multimodal Sentiment Analysis
Chunlei Meng, Jiabin Luo, Zhenglin Yan, Zhenyu Yu, Rong Fu, Zhongxue Gan, Chun Ouyang
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 tri-subspace framework over the best prior method | EMOE · CMU-MOSI (unaligned) | MAE: 0.697 -> 0.691 | Improved |
| compare TSD with ConFEDE on CMU-MOSEI ACC7 | TSD · CMU-MOSEI (unaligned) | ACC7 (%): 54.9 (ConFEDE) vs 54.6 (TSD) | Mixed |
| add submodally shared subspaces (ablation) | TSD · CMU-MOSI (unaligned) | MAE: 0.705 -> 0.691 | Improved |
| add common subspace (ablation) | TSD · CMU-MOSI (unaligned) | MAE: 0.713 -> 0.691 | Improved |
| add private subspaces (ablation) | TSD · CMU-MOSI (unaligned) | MAE: 0.719 -> 0.691 | Improved |
| add tri-subspace disentanglement instead of direct fusion (ablation) | TSD · CMU-MOSI (unaligned) | MAE: 0.722 -> 0.691 | Improved |
| add language modality (ablation) | TSD · CMU-MOSI (unaligned) | MAE: 1.010 -> 0.691 | Improved |
| use SACA fusion instead of summation (ablation) | TSD · CMU-MOSI (unaligned) | MAE: 0.713 -> 0.691 | Improved |
| use SACA instead of CMAF fusion (ablation) | TSD · CMU-MOSEI (unaligned) | MAE: 0.533 -> 0.525 | Improved |
| add common consistency loss Lcom (ablation) | TSD · CMU-MOSI (unaligned) | MAE: 0.719 -> 0.691 | Improved |
| add decoupling supervisor loss Lsup (ablation) | TSD · CMU-MOSI (unaligned) | MAE: 0.721 -> 0.691 | Improved |
| add orthogonality loss Lort (ablation) | TSD · CMU-MOSI (unaligned) | MAE: 0.708 -> 0.691 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- EMOE — EMOE: Modality-Specific Enhanced Dynamic Emotion Experts (CVPR 2025) · better on 10 of 10 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
- DLF: Disentangled-Language-Focused Multimodal Sentiment Analysis (AAAI 2025 · similarity 0.91)
- Toward Robust Incomplete Multimodal Sentiment Analysis via Hierarchical Representation Learning (NeurIPS 2024 · similarity 0.90)
- SemMSA: Latent Semantic-Aided Robust Multimodal Sentiment Analysis with Incomplete Data (NeurIPS 2026 · similarity 0.89)
- A Unified Self-Distillation Framework for Multimodal Sentiment Analysis with Uncertain Missing Modalities (AAAI 2024 · similarity 0.89)
- Enriching Multimodal Sentiment Analysis Through Textual Emotional Descriptions of Visual-Audio Content (AAAI 2025 · similarity 0.88)
- A Multi-Focus-Driven Multi-Branch Network for Robust Multimodal Sentiment Analysis (AAAI 2025 · 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.
