{"ID":6620433,"CreatedAt":"2026-07-15T01:01:48.440468303Z","UpdatedAt":"2026-07-15T03:28:55.185153975Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.12255","arxiv_id":"2607.12255","title":"Understanding Structured Health Data through Interaction-Aware Mixture-of-Experts","abstract":"We study interaction-aware mixture-of-experts for post-stroke rigidity prediction using multi-level views of structured health records. Despite minimal performance gains, routing attribution reveals systematic importance differences across views, underscoring view construction as key to interpretability.","short_abstract":"We study interaction-aware mixture-of-experts for post-stroke rigidity prediction using multi-level views of structured health records. Despite minimal performance gains, routing attribution reveals systematic importance differences across views, underscoring view construction as key to interpretability.","url_abs":"https://arxiv.org/abs/2607.12255","url_pdf":"https://arxiv.org/pdf/2607.12255v1","authors":"[\"Ji Hwan Park\",\"Ying Ding\",\"Tianjin Guo\"]","published":"2026-07-14T01:45:56Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[]","has_code":false}