{"ID":2859623,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.06505","arxiv_id":"2510.06505","title":"A Median Perspective on Unlabeled Data for Out-of-Distribution Detection","abstract":"Out-of-distribution (OOD) detection plays a crucial role in ensuring the robustness and reliability of machine learning systems deployed in real-world applications. Recent approaches have explored the use of unlabeled data, showing potential for enhancing OOD detection capabilities. However, effectively utilizing unlabeled in-the-wild data remains challenging due to the mixed nature of both in-distribution (InD) and OOD samples. The lack of a distinct set of OOD samples complicates the task of training an optimal OOD classifier. In this work, we introduce Medix, a novel framework designed to identify potential outliers from unlabeled data using the median operation. We use the median because it provides a stable estimate of the central tendency, as an OOD detection mechanism, due to its robustness against noise and outliers. Using these identified outliers, along with labeled InD data, we train a robust OOD classifier. From a theoretical perspective, we derive error bounds that demonstrate Medix achieves a low error rate. Empirical results further substantiate our claims, as Medix outperforms existing methods across the board in open-world settings, confirming the validity of our theoretical insights.","short_abstract":"Out-of-distribution (OOD) detection plays a crucial role in ensuring the robustness and reliability of machine learning systems deployed in real-world applications. Recent approaches have explored the use of unlabeled data, showing potential for enhancing OOD detection capabilities. However, effectively utilizing unlab...","url_abs":"https://arxiv.org/abs/2510.06505","url_pdf":"https://arxiv.org/pdf/2510.06505v1","authors":"[\"Momin Abbas\",\"Ali Falahati\",\"Hossein Goli\",\"Mohammad Mohammadi Amiri\"]","published":"2025-10-07T22:43:57Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\",\"math.OC\",\"stat.ML\"]","methods":"[]","has_code":false}