# Robust Conformal Outlier Detection under Contaminated Reference Data

arXiv: https://arxiv.org/abs/2502.04807 · HTML: https://mlsanity.com/paper/2502.04807/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Feb 2025.

## Built on by (most cited first, 12 of 12)

- [General Synthetic-Powered Inference](https://mlsanity.com/paper/2509.20345/) (2025)
- [Conformal Prediction with Corrupted Labels: Uncertain Imputation and Robust Re-weighting](https://mlsanity.com/paper/2505.04733/) (2025)
- [Model-agnostic Selective Labeling with Provable Statistical Guarantees](https://mlsanity.com/paper/2510.14581/) (2025)
- [Conformal Data Contamination Tests for In-distribution Data Acquisition](https://mlsanity.com/paper/2507.13835/) (2025)
- [Synthetic-Powered Multiple Testing with FDR Control](https://mlsanity.com/paper/2602.16690/) (2026)
- [Elements of Conformal Prediction](https://mlsanity.com/paper/2603.23923/) (2026)
- [Online Conformal Anomaly Detection with Prediction-Powered Data Acquisition](https://mlsanity.com/paper/2505.01783/) (2025)
- [Conformal Robust Set Estimation](https://mlsanity.com/paper/2604.18441/) (2026)
- [Conformal calibration and look-elsewhere effect in anomaly detection for new-physics searches](https://mlsanity.com/paper/2606.13780/) (2026)
- [Self-Poisoning in Adaptive Out-of-Distribution Detection: A Sharp-Threshold Theory and Certified Label-Free Calibration](https://mlsanity.com/paper/2607.21673/) (2026)
- [When Modalities Fail to Tango: Conformal Backdoor Detection in Multimodal Contrastive Learning](https://mlsanity.com/paper/2608.04052/) (2026)
- [When Identical Rows Disagree: From Benchmark Identifiability to Replication-Robust Anomaly Detection](https://mlsanity.com/paper/2609.29580/) (2026)

## Abstract (arXiv)

Conformal prediction is a flexible framework for calibrating machine learning predictions, providing distribution-free statistical guarantees. In outlier detection, this calibration relies on a reference set of labeled inlier data to control the type-I error rate. However, obtaining a perfectly labeled inlier reference set is often unrealistic, and a more practical scenario involves access to a contaminated reference set containing a small fraction of outliers. This paper analyzes the impact of such contamination on the validity of conformal methods. We prove that under realistic, non-adversarial settings, calibration on contaminated data yields conservative type-I error control, shedding light on the inherent robustness of conformal methods. This conservativeness, however, typically results in a loss of power. To alleviate this limitation, we propose a novel, active data-cleaning framework that leverages a limited labeling budget and an outlier detection model to selectively annotate data points in the contaminated reference set that are suspected as outliers. By removing only the annotated outliers in this ``suspicious'' subset, we can effectively enhance power while mitigating the risk of inflating the type-I error rate, as supported by our theoretical analysis. Experiments on real datasets validate the conservative behavior of conformal methods under contamination and show that the proposed data-cleaning strategy improves power without sacrificing validity.