Research map / Vision-language and multimodal / Cross-modal learning
Provable Dynamic Fusion for Low-Quality Multimodal Data
Qingyang Zhang, Haitao Wu, Changqing Zhang, Qinghua Hu, Huazhu Fu, Joey Tianyi Zhou, Xi Peng
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 |
|---|---|---|---|
| Use QMF instead of TMC | multimodal classification · NYU Depth V2, Gaussian noise ε=10.0 | average accuracy: 53.36 -> 55.60 | Improved |
| Use QMF instead of Late fusion | multimodal classification · SUN RGB-D, Gaussian noise ε=10.0 | average accuracy: 47.33 -> 48.58 | Improved |
| Use QMF instead of TMC | multimodal classification · FOOD-101, Gaussian noise ε=10.0 | average accuracy: 61.37 -> 62.21 | Improved |
| Use QMF instead of TMC | multimodal classification · MVSA, Gaussian noise ε=10.0 | average accuracy: 60.36 -> 61.28 | Improved |
| Add regularization term (ablation) | QMF without UAW or regularization · NYU Depth V2, Gaussian noise ε=10.0 | average accuracy: 51.94 -> 55.44 | Improved |
| Add uncertainty-aware weighting (ablation) | QMF without UAW or regularization · NYU Depth V2, Gaussian noise ε=10.0 | average accuracy: 51.94 -> 55.14 | Improved |
| Add regularization term (ablation) | QMF with UAW only · NYU Depth V2, Gaussian noise ε=10.0 | average accuracy: 55.14 -> 55.81 | Improved |
| Add uncertainty-aware weighting (ablation) | QMF with regularization only · NYU Depth V2, Gaussian noise ε=10.0 | average accuracy: 55.44 -> 55.81 | Improved |
| Use QMF instead of TMC | multimodal classification · NYU Depth V2, Gaussian noise ε=0.0 | average accuracy: 71.06 -> 70.09 | Hurt |
| Use QMF instead of MMBT | multimodal classification · MVSA, Gaussian noise ε=0.0 | average accuracy: 78.50 -> 78.07 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- MMTM — MMTM: Multimodal Transfer Module for CNN Fusion (CVPR 2020) · better on 0 of 1 shared result
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
- Predictive Dynamic Fusion (ICML 2024 · similarity 0.91)
- Embracing Unimodal Aleatoric Uncertainty for Robust Multimodal Fusion (CVPR 2024 · similarity 0.89)
- Modality-Aware SAM: Sharpness-Aware-Minimization Driven Gradient Modulation for Harmonized Multimodal Learning (NeurIPS 2025 · similarity 0.87)
- Fusion or Confusion? Multimodal Complexity Is Not All You Need (NeurIPS 2026 · similarity 0.87)
- Amplifying Prominent Representations in Multimodal Learning via Variational Dirichlet Process (NeurIPS 2025 · similarity 0.87)
- Beyond Modality Fusion: Deep Ensembles for Multimodal Classification (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.
