aiqbench analysis
This research introduces a scalable post-hoc method for out-of-distribution detection in machine learning systems by leveraging counterfactual explanations to measure distance from decision boundaries. By computing these metrics directly in embedding space, the approach significantly reduces computational costs while maintaining state-of-the-art performance on standard benchmarks like CIFAR-100 and ImageNet-200. The integration of interpretability allows practitioners to not only detect anomalies but also understand the specific reasons behind detection failures, enhancing the safety and trustworthiness of deployed AI models.
Out-of-Distribution Detection using Counterfactual Distance
arXiv:2508.10148v2 Announce Type: replace-cross Abstract: Accurate and explainable out-of-distribution (OOD) detection is required to use machine learning systems safely. Previous work has shown that feature distance to decision boundaries can be used to identify OOD data effectively. In this paper, we build on this intuition and propose a post-hoc OOD detection method that, given an input, calculates the distance to decision boundaries by leveraging counterfactual explanations. Since computing explanations can be expensive for large architectures, we also propose strategies to improve scalability by computing counterfactuals directly in embedding space. Crucially, as the method employs counterfactual explanations, we can seamlessly use them to help interpret the results of our detector. We show that our method is in line with the state of the art on CIFAR-10, achieving 93.50% AUROC and 25.80% FPR95. Our method outperforms these methods on CIFAR-100 with 97.05% AUROC and 13.79% FPR95 and on ImageNet-200 with 92.55% AUROC and 33.55% FPR95 across four OOD datasets
Abstract:Accurate and explainable out-of-distribution (OOD) detection is required to use machine learning systems safely. Previous work has shown that feature distance to decision boundaries can be used to identify OOD data effectively. In this paper, we build on this intuition and propose a post-hoc OOD detection method that, given an input, calculates the distance to decision boundaries by leveraging counterfactual explanations. Since computing explanations can be expensive for large architectures, we also propose strategies to improve scalability by computing counterfactuals directly in embedding space. Crucially, as the method employs counterfactual explanations, we can seamlessly use them to help interpret the results of our detector. We show that our method is in line with the state of the art on CIFAR-10, achieving 93.50% AUROC and 25.80% FPR95. Our method outperforms these methods on CIFAR-100 with 97.05% AUROC and 13.79% FPR95 and on ImageNet-200 with 92.55% AUROC and 33.55% FPR95 across four OOD datasets
Presented at the IJCAI 2026 Workshop on Explainable Artificial Intelligence (XAI). 11 pages, 3 figures, 4 tables
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2508.10148 [cs.LG]
(or arXiv:2508.10148v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2508.10148 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 264 words
Related coverage
Coherence-Aware Distributional Evaluation of Open-Ended Text Generation
arXiv · Sep 29, 2026
Why do models task game?
AI Alignment Forum · Aug 6, 2026
How bad would it be if GPS satellites were shot down?
LessWrong · Jun 12, 2026
Inference meta-monitoring for Amazon SageMaker AI endpoints with Amazon Quick
AWS · Jul 30, 2026