Research map / Vision-language and multimodal / Cross-modal learning
SimMLM: A Simple Framework for Multi-modal Learning with Missing Modality
Sijie Li, Chen Chen, Jungong Han
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 MoFe ranking loss (ablation) | DMoME · BraTS 2018 test (ET) | Dice score: 66.05 -> 67.16 | Improved |
| add MoFe ranking loss (ablation) | DMoME · BraTS 2018 test (TC) | Dice score: 79.14 -> 80.20 | Improved |
| add MoFe ranking loss (ablation) | DMoME · BraTS 2018 test (WT) | Dice score: 86.71 -> 87.67 | Improved |
| add expert pretraining stage (ablation) | DMoME · BraTS 2018 validation (ET) | Dice score: 62.24 -> 63.22 | Improved |
| use dynamic gating instead of static averaging (ablation) | DMoME · BraTS 2018 validation (ET) | Dice score: 61.40 -> 63.22 | Improved |
| weight experts at logit level instead of feature level (ablation) | DMoME · BraTS 2018 validation (ET) | Dice score: 62.76 -> 63.22 | Improved |
| weight experts at logit level instead of probability level (ablation) | DMoME · BraTS 2018 validation (WT) | Dice score: 87.33 -> 87.21 | No clear effect |
| use MoFe loss instead of confidence ranking loss (ablation) | DMoME · BraTS 2018 validation (ET) | ECE (%): 3.17 -> 3.15 | Improved |
| add MoFe ranking loss (ablation) | DMoME · BraTS 2018 test (ET) | counterintuitive rate (%): 10.47 -> 7.20 | Improved |
| use DMoME instead of MoMKE (dedicated expert per modality) | MoMKE · BraTS 2018 test | FLOPS: 490G -> 123G | Improved |
| use DMoME w/o MoFe instead of MoMKE | MoMKE · BraTS 2018 test (ET) | Dice score: 65.56 -> 66.05 | Improved |
| vary MoFe coefficient lambda in {0, 0.01, 0.05, 0.1, 0.2, 0.5, 1} (ablation) | SimMLM · BraTS 2018 validation | Dice score: best at lambda=0.1 | Improved |
Later papers that compared with this method
1 later accepted paper reports SimMLM as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Residual-Guided Expert Specialization for Incomplete Multimodal Learning (ECCV 2026 · better on 1 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
- Flex-MoE: Modeling Arbitrary Modality Combination via the Flexible Mixture-of-Experts (NeurIPS 2024 · similarity 0.89)
- Do More Modalities Always Help? A Geometric Perspective on Missing-Modality Robustness (NeurIPS 2026 · similarity 0.88)
- Gradient-Guided Modality Decoupling for Missing-Modality Robustness (AAAI 2024 · similarity 0.88)
- Knowledge Bridger: Towards Training-Free Missing Modality Completion (CVPR 2025 · similarity 0.87)
- Jointly Modeling Inter- & Intra-Modality Dependencies for Multi-modal Learning (NeurIPS 2024 · similarity 0.87)
- RedCore: Relative Advantage Aware Cross-Modal Representation Learning for Missing Modalities with Imbalanced Missing Rates (AAAI 2024 · similarity 0.87)
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.
