Metraipolis › archive › dc_2d6a7a4aeade

Why Is SHAP Not a Reliable Standalone Explanation Framework for Malware Detection? — arXiv:2609.04626

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.04626
License (as archived)
CC0-1.0
Archived
2026-10-16T22:51:15.955Z
SHA-256 of the archived text
3fccb5d9d7baaa2216cd42e386e954c7514425e807c0e87b6816ace3e779b587
Stable URL of this page
https://metraipolis.com/documento/dc_2d6a7a4aeade

How to cite

Why Is SHAP Not a Reliable Standalone Explanation Framework for Malware Detection? — arXiv:2609.04626 — arXiv (en), https://arxiv.org/abs/2609.04626 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2d6a7a4aeade (Metraipolis archive dc_2d6a7a4aeade, sha256 3fccb5d9d7baaa2216cd42e386e954c7514425e807c0e87b6816ace3e779b587)
BibTeX
@misc{dc2d6a7a4aeade,
  title = {Why Is SHAP Not a Reliable Standalone Explanation Framework for Malware Detection? — arXiv:2609.04626},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.04626},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2d6a7a4aeade; sha256 3fccb5d9d7baaa2216cd42e386e954c7514425e807c0e87b6816ace3e779b587},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04626. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Machine learning is widely used for malware detection, but its decisions must be explained. An analyst needs to know whether a model has learned genuine malicious behavior or only dataset-specific patterns \cite{gaur2021semantics}. SHapley Additive exPlanations (SHAP) is the standard tool for this, backed by formal properties such as local accuracy, missingness, and consistency. We argue that these guarantees are insufficient for reliable malware interpretation. We claim SHAP explains a chosen feature-coalition game, not malware behavior in the data. That game is fixed only after the analyst selects the feature players, the missing feature rule, the background distribution, and the simplified input mapping. In static Portable Executable feature spaces, groups such as byte histograms, byte-entropy, strings, headers, sections, imports, and data-directories are not independent signals but are jointly shaped by file structure, packing, compiler behavior, and family conventions. We prove that this dependence makes conditional SHAP dilute a model's feature credit by a factor of $1/m$ across $m-1$ redundant features, attributes importance to features the model never uses, and even reverses the sign of an unused feature's attribution when the data distribution changes; interventional SHAP, meanwhile, queries off-manifold coalitions that no real executable would exhibit. Experiments on EMBER-2018, EMBER-2024, and BODMAS with fixed LightGBM and XGBoost detectors confirm these effects. We therefore position SHAP as a limited diagnostic that requires an explicitly stated data distribution and domain validation, not a standalone account of malware behavior.

Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-07 in cs.CR).
Categorie arXiv: Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2609.04626 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T22:30:54.308Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).