Research map / Vision-language and multimodal / Cross-modal learning
Multimodal Learning on Low-Quality Data with Conformal Predictive Self-Calibration
Xun Jiang, Yufan Gu, Disen Hu, Yuqing Hou, Yazhou Yao, Fumin Shen, Heng Tao Shen, Xing Xu
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 RSC to GSC-only model (ablation) | CPSC · CREMA-D | multimodal accuracy: 86.29 -> 87.83 | Improved |
| add GSC to RSC-only model (ablation) | CPSC · CREMA-D | multimodal accuracy: 86.35 -> 87.83 | Improved |
| add GSC to the plain baseline (ablation) | baseline · CREMA-D | multimodal accuracy: 78.23 -> 86.29 | Improved |
| add RSC to the plain baseline (ablation) | baseline · CREMA-D | multimodal accuracy: 78.23 -> 86.35 | Improved |
| add RSC to the plain baseline (ablation) | baseline · NYU Depth V2 (clean) | accuracy: 69.16 -> 71.36 | Improved |
| add GSC alone to the plain baseline (ablation) | baseline · NYU Depth V2 (clean) | accuracy: 69.16 -> 70.14 | Improved |
| add GSC alone to the plain baseline under strong Gaussian noise (ablation) | baseline · NYU Depth V2 (Gaussian noise eps=10.0) | accuracy: 51.99 -> 50.17 | Hurt |
| add GSC to RSC-only model under strong Gaussian noise (ablation) | CPSC · NYU Depth V2 (Gaussian noise eps=10.0) | accuracy: 53.82 -> 57.32 | Improved |
| add the method to Adam optimizer | Adam · CREMA-D | multimodal accuracy: 83.78 -> 85.29 | Improved |
| add the method to AdaGrad optimizer | AdaGrad · CREMA-D | multimodal accuracy: 68.31 -> 76.68 | Improved |
| add the method to Adam optimizer | Adam · Kinetics Sounds | multimodal accuracy: 72.18 -> 75.13 | Improved |
| add diversity and consistency constraints to RSC (ablation) | RSC · CREMA-D and NYU Depth V2 | accuracy | Improved |
Closest papers by meaning
- Self-Calibrating Conformal Prediction (NeurIPS 2024 · similarity 0.88)
- Asymmetric Reinforcing Against Multi-Modal Representation Bias (AAAI 2025 · similarity 0.86)
- When to Align, When to Predict: A Phase Diagram for Multimodal Learning (NeurIPS 2026 · similarity 0.86)
- Multimodal Negative Learning (NeurIPS 2025 · similarity 0.86)
- MIDAS: Misalignment-based Data Augmentation Strategy for Imbalanced Multimodal Learning (NeurIPS 2025 · similarity 0.86)
- Fusion or Confusion? Multimodal Complexity Is Not All You Need (NeurIPS 2026 · similarity 0.85)
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.
