# Conformal Prediction and Conditional Coverage for Tabular Foundation Models

arXiv: https://arxiv.org/abs/2609.34887 · HTML: https://mlsanity.com/paper/2609.34887/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 22 references that are in the map.
- First posted on arXiv 28 Sep 2026.

## Builds on (most cited first, 13 of 13)

- [Conformalized Quantile Regression](https://mlsanity.com/paper/1905.03222/) (2019)
- [TabICLv2: A better, faster, scalable, and open tabular foundation model](https://mlsanity.com/paper/2602.11139/) (2026)
- [Learning by Transduction](https://mlsanity.com/paper/1301.7375/) (2013)
- [Conformal Prediction using Conditional Histograms](https://mlsanity.com/paper/2105.08747/) (2021)
- [CD-split and HPD-split: efficient conformal regions in high dimensions](https://mlsanity.com/paper/2007.12778/) (2020)
- [CARTE: Pretraining and Transfer for Tabular Learning](https://mlsanity.com/paper/2402.16785/) (2024)
- [A Unified Comparative Study with Generalized Conformity Scores for Multi-Output Conformal Regression](https://mlsanity.com/paper/2501.10533/) (2025)
- [Bayes' Power for Explaining In-Context Learning Generalizations](https://mlsanity.com/paper/2410.01565/) (2024)
- [Towards Fair In-Context Learning with Tabular Foundation Models](https://mlsanity.com/paper/2505.09503/) (2025)
- [JAPAN: Joint Adaptive Prediction Areas with Normalising-Flows](https://mlsanity.com/paper/2505.23196/) (2025)
- [A Post-Processing Conformal Prediction Approach for Conditional Coverage via Pivotal Scores](https://mlsanity.com/paper/2605.25852/) (2026)
- [High Performance, Low Reliability: Uncertainty Benchmarking for Tabular Foundation Models](https://mlsanity.com/paper/2605.28554/) (2026)
- [On Optimal Data Splitting for Split Conformal Prediction](https://mlsanity.com/paper/2606.31600/) (2026)

## Abstract (arXiv)

Tabular foundation models (TFMs) provide predictive distributions for regression, but their prediction regions can exhibit undercoverage or overcoverage even when point predictions are accurate. We introduce C-USIM (Conditionally-Uniformized Score Integration Method), a lightweight application of highest predictive density split conformal prediction that accommodates multimodal predictions. Given calibration and test outputs, it requires no additional training or model inference. It provides finite-sample marginal validity under our assumptions. We bound conditional-marginal coverage gaps using distribution-estimation error and score discreteness, and examine coverage heterogeneity through percentile rank-score plots. Experiments with TabPFN and TabICL show improved marginal coverage accuracy and lower average conditional and group coverage errors. Under a fixed data budget, allocating more observations to calibration can reduce marginal coverage error despite less accurate point predictions.