Research map / Label-efficient and robust learning / Self-supervised and contrastive
Unleashing the Power of Image-Tabular Self-Supervised Learning via Breaking Cross-Tabular Barriers
Yibing Fu, Yunpeng Zhao, Zhitao Zeng, Cheng Chen, Yueming Jin
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 |
|---|---|---|---|
| remove column-header semantic embedding (ablation) | CITab · ADNI | accuracy: 0.654 -> 0.846 | Improved |
| remove column-header semantic embedding (ablation) | CITab · AIBL / NACC | accuracy: 0.752 -> 0.817 (AIBL); 0.690 -> 0.728 (NACC) | Improved |
| add image-tabular contrastive loss in pretraining (ablation) | CITab · ADNI | accuracy: 0.735 -> 0.846 | Improved |
| add image-tabular matching loss (ablation) | CITab · ADNI | accuracy: 0.799 -> 0.846 | Improved |
| add multi-modal SSL pretraining across cohorts (ablation) | CITab · AIBL | accuracy: 0.756 -> 0.817 | Improved |
| add P-MoLin module (ablation) | CITab · NACC | accuracy: 0.733 -> 0.728 | Hurt |
| add P-MoLin module (ablation) | CITab · ADNI | accuracy: 0.823 -> 0.846 | Improved |
| use unified self-attention rather than cross-attention fusion (ablation) | CITab · AIBL | accuracy: 0.744 -> 0.817 | Improved |
| use general BERT rather than BioBERT for headers (ablation) | CITab · AIBL | accuracy: 0.736 -> 0.817 | Improved |
| increase P-MoLin experts from 5 to 6 (ablation) | CITab · 3-cohort average | accuracy: 0.797 -> 0.787 | Hurt |
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.
- ViTa — VITA: A Multi-Source Vicinal Transfer Augmentation Method for Out-of-Distribution Generalization (AAAI 2022) · better on 3 of 3 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
- ConTextTab: A Semantics-Aware Tabular In-Context Learner (NeurIPS 2025 · similarity 0.89)
- Small Models are LLM Knowledge Triggers for Medical Tabular Prediction (ICLR 2025 · similarity 0.88)
- Continual Self-supervised Learning: Towards Universal Multi-modal Medical Data Representation Learning (CVPR 2024 · similarity 0.88)
- SwitchTab: Switched Autoencoders Are Effective Tabular Learners (AAAI 2024 · similarity 0.88)
- TabGLM: Tabular Graph Language Model for Learning Transferable Representations Through Multi-Modal Consistency Minimization (AAAI 2025 · similarity 0.87)
- Large Scale Transfer Learning for Tabular Data via Language Modeling (NeurIPS 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.
