Research map / Vision-language and multimodal / Cross-modal learning
Modality to Modality Translation: An Adversarial Representation Learning and Graph Fusion Network for Multimodal Fusion
Sijie Mai, Haifeng Hu, Songlong Xing
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 adversarial training to align modality distributions (ablation) | ARGF · CMU-MOSI | Accuracy: 80.11 -> 81.38 | Improved |
| add adversarial training to align modality distributions (ablation) | ARGF · CMU-MOSI | F1 score: 80.23 -> 81.52 | Improved |
| add the classifier to the embedding-space learning stage (ablation) | ARGF · CMU-MOSI | Accuracy: 80.92 -> 81.38 | Improved |
| add the classifier to the embedding-space learning stage (ablation) | ARGF · CMU-MOSI | F1 score: 80.98 -> 81.52 | Improved |
| add decoders to preserve modality-specific information (ablation) | ARGF · CMU-MOSI | Accuracy: 79.72 -> 81.38 | Improved |
| add decoders to preserve modality-specific information (ablation) | ARGF · CMU-MOSI | F1 score: 79.76 -> 81.52 | Improved |
| use hierarchical Graph Fusion Network instead of alternative fusion structures | ARGF · CMU-MOSI | Accuracy: 79.92 -> 81.38 | Improved |
| use hierarchical Graph Fusion Network instead of alternative fusion structures | ARGF · IEMOCAP | Accuracy: 57.79 -> 60.20 | Improved |
| choose a different modality as the target distribution | ARGF · CMU-MOSI | Avg Accuracy: 81.38; 81.25; 81.12 | No clear effect |
Closest papers by meaning
- A Closer Look at Multimodal Representation Collapse (ICML 2025 · similarity 0.90)
- Beyond Modality Fusion: Deep Ensembles for Multimodal Classification (NeurIPS 2026 · similarity 0.89)
- Modality-Aware SAM: Sharpness-Aware-Minimization Driven Gradient Modulation for Harmonized Multimodal Learning (NeurIPS 2025 · similarity 0.87)
- Connect, Collapse, Corrupt: Learning Cross-Modal Tasks with Uni-Modal Data (ICLR 2024 · similarity 0.87)
- Embracing Unimodal Aleatoric Uncertainty for Robust Multimodal Fusion (CVPR 2024 · similarity 0.87)
- MSP: Modality Self-Play for Training Multimodal LLMs Without Paired Cross-Modal Data (NeurIPS 2026 · similarity 0.86)
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.
