{"paper":{"title":"Delta-Audit: Explaining What Changes When Models Change","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Afsaneh Fatemi, Arshia Hemmat","submitted_at":"2025-08-27T05:52:38Z","abstract_excerpt":"Model updates (new hyperparameters, kernels, depths, solvers, or data) change performance, but the \\emph{reason} often remains opaque. We introduce \\textbf{Delta-Attribution} (\\mbox{$\\Delta$-Attribution}), a model-agnostic framework that explains \\emph{what changed} between versions $A$ and $B$ by differencing per-feature attributions: $\\Delta\\phi(x)=\\phi_B(x)-\\phi_A(x)$. We evaluate $\\Delta\\phi$ with a \\emph{$\\Delta$-Attribution Quality Suite} covering magnitude/sparsity (L1, Top-$k$, entropy), agreement/shift (rank-overlap@10, Jensen--Shannon divergence), behavioural alignment (Delta Conserv"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.19589","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2508.19589/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}