Research map / Probabilistic and scientific modeling / Molecules and proteins
Using maximal information auxiliary variables to improve synthetic data generation based on TabPFN foundation models
Elias Chaibub Neto
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 |
|---|---|---|---|
| replace direct TabPFN generation (JF) with MIAV auxiliary variables | TabPFN synthetic data generation · simulated, real OpenML-CC18 (36) and 7 baseline datasets | fidelity (KS, L2D, DT) | Improved |
| weaken variable association strength (rho toward 0) for direct JF and FC strategies (ablation) | JF and FC TabPFN generators · simulated correlated beta data | marginal distribution quality | Hurt |
| use MIAV instead of SMOTE | synthetic data generator · real datasets | fidelity | Hurt |
| use MIAV instead of SMOTE | synthetic data generator · real datasets | privacy (DCR, SDBRL, SSDID) | Improved |
| use MIAV instead of DDPM, CTGAN, TVAE, ARF, BN | synthetic data generator · 7 real datasets | fidelity | Mixed |
| use MIAV instead of JF and FC | synthetic data generator · real datasets | privacy | Hurt |
| build MIAV on TabICL instead of TabPFN (ablation) | MIAV · 8 categorical OpenML-CC18 datasets | synthetic data quality | No clear effect |
Closest papers by meaning
- Tabular Foundation Model for Generative Modelling (NeurIPS 2026 · similarity 0.86)
- Towards a Theoretical Understanding of Synthetic Data in LLM Post-Training: A Reverse-Bottleneck Perspective (ICLR 2025 · similarity 0.86)
- Drift-Resilient TabPFN: In-Context Learning Temporal Distribution Shifts on Tabular Data (NeurIPS 2024 · similarity 0.86)
- How Realistic Is Your Synthetic Data? Constraining Deep Generative Models for Tabular Data (ICLR 2024 · similarity 0.86)
- TabDPT: Scaling Tabular Foundation Models on Real Data (NeurIPS 2025 · similarity 0.86)
- TSGM: A Flexible Framework for Generative Modeling of Synthetic Time Series (NeurIPS 2024 · 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.
