A principled framework for uncertainty decomposition in TabPFN

2026-02-04 · Sandra Fortini, Kenyon Ng, Sonia Petrone, Judith Rousseau, Susan Wei arxiv

TabPFN is a transformer that achieves state-of-the-art performance on supervised tabular tasks by amortizing Bayesian prediction into a single forward pass. However, there is currently no method for uncertainty decomposition in TabPFN. Because it behaves, in an idealised limit, as a Bayesian in-context learner, we cast the decomposition challenge as a Bayesian predictive inference (BPI) problem. The main computational tool in BPI, predictive Monte Carlo, is challenging to apply here as it requires simulating unmodeled covariates. We therefore pursue the asymptotic alternative, filling a gap in the theory for supervised settings by proving a predictive CLT under quasi-martingale conditions. We derive variance estimators determined by the volatility of predictive updates along the context. The resulting credible bands are fast to compute, target epistemic uncertainty, and achieve near-nominal frequentist coverage. For classification, we further obtain an entropy-based uncertainty decomposition.

📄 PDF Abstract BibTeX arXiv:2602.04596

Code (0)

등록된 구현이 없습니다.

TabMGP: Martingale Posterior with TabPFN

2025-10-29 · Kenyon Ng, Edwin Fong, David T. Frazier, Jeremias Knoblauch 외 arxiv

Bayesian inference provides principled uncertainty quantification but is often limited by the challenges of prior and likelihood elicitation. The martingale posterior (MGP) (Fong et al., 2023) offers an alternative by re…

Bayesian Inference

uLEAD-TabPFN: Uncertainty-aware Dependency-based Anomaly Detection with TabPFN

2026-04-22 · Sha Lu, Jixue Liu, Stefan Peters, Thuc Duy Le 외 arxiv

Anomaly detection in tabular data is challenging due to high dimensionality, complex feature dependencies, and heterogeneous noise. Many existing methods rely on proximity-based cues and may miss anomalies caused by viol…

Anomaly Detection

TabPFN Extensions for Interpretable Geotechnical Modelling

2026-03-22 · Taiga Saito, Yu Otake, Daijiro Mizutani, Stephen Wu arxiv

Geotechnical site characterisation relies on sparse, heterogeneous borehole data, where uncertainty quantification and interpretability matter as much as predictive accuracy. We evaluate TabPFN~\citep{Hollmann2025}, a ta…

Uncertainty-Aware Tabular Prediction: Evaluating VBLL-Enhanced TabPFN in Safety-Critical Medical Data

2025-09-12 · Madhushan Ramalingam arxiv

Predictive models are being increasingly used across a wide range of domains, including safety-critical applications such as medical diagnosis and criminal justice. Reliable uncertainty estimation is a crucial task in su…

Medical Diagnosis

Foundation-Model Surrogates Enable Data-Efficient Active Learning for Materials Discovery

2026-03-13 · Jeffrey Hu, Rongzhi Dong, Ying Feng, Ming Hu 외 arxiv

Active learning (AL) has emerged as a powerful paradigm for accelerating materials discovery by iteratively steering experiments toward promising candidates, reducing the number of costly synthesis-and-characterization c…

Bayesian Inference Active Learning