MARVEL: Margin-Aware Robust von Mises-Fischer Expert Learning for Long-Tailed Out-of-Distribution Detection

• URL: http://arxiv.org/abs/2607.02435v1
• Date: Thu, 02 Jul 2026 17:06:31 GMT
• Title: MARVEL: Margin-Aware Robust von Mises-Fischer Expert Learning for Long-Tailed Out-of-Distribution Detection
• Abstract summary: Deep models routinely misclassify out-of-distribution (OOD) inputs with high confidence.<n>Most of the existing literature assumes balanced datasets.<n>We propose a novel methodology trained on diverse imbalanced medical datasets.
• Score: 0.8886706641070187
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: For clinical deployment, it is essential that automated diagnostic systems remain reliable when confronted with previously unseen cases, yet deep models routinely misclassify out-of-distribution (OOD) inputs with high confidence, underscoring the need for more robust OOD detection methods. Although substantial effort has been devoted to improving model robustness, most of the existing literature assumes balanced datasets, evaluates OOD detection on coarse or non-clinical OOD sources, or lacks comprehensive assessment across diverse OOD scenarios. To address the gaps, we propose a novel methodology trained on diverse and imbalanced medical datasets and evaluated across a clinically reflective OOD spectrum. Our framework comprises three key components: (1) a Nonlinear von Mises-Fisher (NvMF) classifier capable of learning non-linear decision boundaries, with theoretical proof of its asymptotic connection to cosine classifiers; (2) a multi-expert framework in which margin-aware NvMF classifiers specialise in different regions of label distribution to better handle imbalance; and (3) an outlier expert trained explicitly to distinguish inlier from outlier data, thereby strengthening OOD detection. Evaluation on RFMiD, ISIC2019, and NCTCRC datasets demonstrates consistent improvements over state-of-the-art methods, achieving mean FPR95 reductions of 8.45%, 13.02%, and 36.90% respectively. These gains are further supported by comprehensive ablations that validated the contributions of each component. This enables reliable identification of unfamiliar cases for deferral to clinicians, supporting safer AI-assisted diagnosis in real-world workflows. Our code is available at https://github.com/redboxup/MARVEL.

Related papers

• ProOOD: Prototype-Guided Out-of-Distribution 3D Occupancy Prediction [77.70461178466043]
  3D semantic occupancy prediction is central to autonomous driving.<n>Current methods are vulnerable to long-tailed class bias and out-of-distribution (OOD) inputs.<n>We present ProOOD, a lightweight, plug-and-play method that couples prototype-guided refinement with training-free OOD scoring.
  arXiv  Detail & Related papers  (2026-04-01T16:11:59Z)
• Revisiting Logit Distributions for Reliable Out-of-Distribution Detection [73.9121001113687]
  Out-of-distribution (OOD) detection is critical for ensuring the reliability of deep learning models in open-world applications.<n>LogitGap is a novel post-hoc OOD detection method that exploits the relationship between the maximum logit and the remaining logits.<n>We show that LogitGap consistently achieves state-of-the-art performance across diverse OOD detection scenarios and benchmarks.
  arXiv  Detail & Related papers  (2025-10-23T02:16:45Z)
• A Median Perspective on Unlabeled Data for Out-of-Distribution Detection [5.937613452723966]
  Out-of-distribution (OOD) detection plays a crucial role in ensuring the robustness and reliability of machine learning systems.<n>Recent approaches have explored the use of unlabeled data, showing potential for enhancing OOD detection capabilities.<n>We introduce Medix, a novel framework designed to identify potential outliers from unlabeled data using the median operation.
  arXiv  Detail & Related papers  (2025-10-07T22:43:57Z)
• DCV-ROOD Evaluation Framework: Dual Cross-Validation for Robust Out-of-Distribution Detection [2.9726444682922897]
  Out-of-distribution (OOD) detection plays a key role in enhancing the robustness of artificial intelligence systems.<n>Cross-validation (CV) has proven to be a highly effective tool for providing a reasonable estimate of the performance of a learning algorithm.<n>This work proposes a dual CV framework for robust evaluation of OOD detection models.
  arXiv  Detail & Related papers  (2025-09-06T17:20:09Z)
• NERO: Explainable Out-of-Distribution Detection with Neuron-level Relevance [9.86792165443301]
  We propose a novel OOD scoring mechanism, called NERO, that leverages neuron-level relevance at the feature layer.<n>Specifically, we cluster neuron-level relevance for each in-distribution (ID) class to form representative centroids.<n>We refine performance by incorporating scaled relevance in the bias term and combining feature norms.
  arXiv  Detail & Related papers  (2025-06-18T12:22:17Z)
• CADRef: Robust Out-of-Distribution Detection via Class-Aware Decoupled Relative Feature Leveraging [5.356623181327855]
  Class-Aware Relative Feature-based method (CARef) and Class-Aware Decoupled Relative Feature-based method (CADRef) are proposed.<n>We show that both proposed methods exhibit effectiveness and robustness in OOD detection compared to state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-03-01T03:23:10Z)
• Margin-bounded Confidence Scores for Out-of-Distribution Detection [2.373572816573706]
  We propose a novel method called Margin bounded Confidence Scores (MaCS) to address the nontrivial OOD detection problem. MaCS enlarges the disparity between ID and OOD scores, which in turn makes the decision boundary more compact. Experiments on various benchmark datasets for image classification tasks demonstrate the effectiveness of the proposed method.
  arXiv  Detail & Related papers  (2024-09-22T05:40:25Z)
• How Does Unlabeled Data Provably Help Out-of-Distribution Detection? [63.41681272937562]
  Unlabeled in-the-wild data is non-trivial due to the heterogeneity of both in-distribution (ID) and out-of-distribution (OOD) data. This paper introduces a new learning framework SAL (Separate And Learn) that offers both strong theoretical guarantees and empirical effectiveness.
  arXiv  Detail & Related papers  (2024-02-05T20:36:33Z)
• Conservative Prediction via Data-Driven Confidence Minimization [70.93946578046003]
  In safety-critical applications of machine learning, it is often desirable for a model to be conservative. We propose the Data-Driven Confidence Minimization framework, which minimizes confidence on an uncertainty dataset.
  arXiv  Detail & Related papers  (2023-06-08T07:05:36Z)
• Free Lunch for Generating Effective Outlier Supervision [46.37464572099351]
  We propose an ultra-effective method to generate near-realistic outlier supervision. Our proposed textttBayesAug significantly reduces the false positive rate over 12.50% compared with the previous schemes.
  arXiv  Detail & Related papers  (2023-01-17T01:46:45Z)
• Provably Robust Detection of Out-of-distribution Data (almost) for free [124.14121487542613]
  Deep neural networks are known to produce highly overconfident predictions on out-of-distribution (OOD) data. In this paper we propose a novel method where from first principles we combine a certifiable OOD detector with a standard classifier into an OOD aware classifier. In this way we achieve the best of two worlds: certifiably adversarially robust OOD detection, even for OOD samples close to the in-distribution, without loss in prediction accuracy and close to state-of-the-art OOD detection performance for non-manipulated OOD data.
  arXiv  Detail & Related papers  (2021-06-08T11:40:49Z)
• Robust Out-of-distribution Detection for Neural Networks [51.19164318924997]
  We show that existing detection mechanisms can be extremely brittle when evaluating on in-distribution and OOD inputs. We propose an effective algorithm called ALOE, which performs robust training by exposing the model to both adversarially crafted inlier and outlier examples.
  arXiv  Detail & Related papers  (2020-03-21T17:46:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.