From Behavior to Provenance: Attributing Tabular Foundation Models to Synthetic Pretraining Data

arXiv:2610.02347 · cs.LG · Submitted 2026-10-01 · Read on arXiv

cs.LG

Submitted: 2026-10-01

Updated: 2026-10-01

Terminology

• Foundation Model — 1× in this paper · explained in 3 episodes
• Foundation Models — 1× in this paper · explained in 7 episodes
• Provenance — 1× in this paper · explained in 4 episodes

Sources

• TRAK: Attributing Model Behavior at Scale
• TabPFN-3.5: Technical Report
• TabICLv2: A better, faster, scalable, and open tabular foundation model
• Orion-MSP: Multi-Scale Sparse Attention for Tabular In-Context Learning
• LimiX-2: A Contextual Mechanism Network Towards General Structured-Data Intelligence
• Shaping the Prior: How Synthetic Task Distributions Determine Tabular Foundation Model Quality
• nanoTabPFN: A Lightweight and Educational Reimplementation of TabPFN
• OpenML Benchmarking Suites

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks