{"ID":2893741,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.11855","arxiv_id":"2507.11855","title":"OrdShap: Feature Position Importance for Sequential Black-Box Models","abstract":"Sequential deep learning models excel in domains with temporal or sequential dependencies, but their complexity necessitates post-hoc feature attribution methods for understanding their predictions. While existing techniques quantify feature importance, they inherently assume fixed feature ordering - conflating the effects of (1) feature values and (2) their positions within input sequences. To address this gap, we introduce OrdShap, a novel attribution method that disentangles these effects by quantifying how a model's predictions change in response to permuting feature position. We establish a game-theoretic connection between OrdShap and Sanchez-Bergantiños values, providing a theoretically grounded approach to position-sensitive attribution. Empirical results from health, natural language, and synthetic datasets highlight OrdShap's effectiveness in capturing feature value and feature position attributions, and provide deeper insight into model behavior.","short_abstract":"Sequential deep learning models excel in domains with temporal or sequential dependencies, but their complexity necessitates post-hoc feature attribution methods for understanding their predictions. While existing techniques quantify feature importance, they inherently assume fixed feature ordering - conflating the eff...","url_abs":"https://arxiv.org/abs/2507.11855","url_pdf":"https://arxiv.org/pdf/2507.11855v2","authors":"[\"Davin Hill\",\"Brian L. Hill\",\"Aria Masoomi\",\"Vijay S. Nori\",\"Robert E. Tillman\",\"Jennifer Dy\"]","published":"2025-07-16T02:40:01Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[\"Generative Adversarial Network\"]","has_code":false}