Metraipolis › archive › dc_030d358a7f0f

From Behavior to Provenance: Attributing Tabular Foundation Models to Synthetic Pretraining Data — arXiv:2610.02347

Source: arXiv
Original: https://arxiv.org/abs/2610.02347
License (as archived): CC0-1.0
Archived: 2026-10-05T23:08:49.928Z
SHA-256 of the archived text: 4d66407d4c5fe61178c6fefc86ab42f9b4a275eb4aa3318beb8b505ab38a659e
Stable URL of this page: https://metraipolis.com/documento/dc_030d358a7f0f

How to cite

From Behavior to Provenance: Attributing Tabular Foundation Models to Synthetic Pretraining Data — arXiv:2610.02347 — arXiv, https://arxiv.org/abs/2610.02347 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_030d358a7f0f (Metraipolis archive dc_030d358a7f0f, sha256 4d66407d4c5fe61178c6fefc86ab42f9b4a275eb4aa3318beb8b505ab38a659e)

BibTeX

@misc{dc030d358a7f0f,
  title = {From Behavior to Provenance: Attributing Tabular Foundation Models to Synthetic Pretraining Data — arXiv:2610.02347},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.02347},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_030d358a7f0f; sha256 4d66407d4c5fe61178c6fefc86ab42f9b4a275eb4aa3318beb8b505ab38a659e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.02347. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Training-data attribution aims to identify which training examples shape model behavior, yet validating such claims is difficult because causal training influence is rarely observable. We argue that controlled synthetic pretraining makes attribution experimentally testable. Using O'PRIOR, a provenance-rich synthetic task generator for tabular foundation models, we construct a testbed in which every pretraining task carries explicit lineage over structural mechanisms, missingness, confounding, shortcuts, and distribution shift. We combine behavior-conditioned attribution with counterfactual retraining and provenance-aware interventions to test both task-level faithfulness and mechanism-level consistency. On held-out real tasks, removing the top-attributed 5% of synthetic tasks decreases mean ROC-AUC by 0.013, compared with 0.002$\pm$0.004 under random removal, while removing bottom-attributed tasks improves performance by 0.003. Within shortcut-provenance tasks, targeted removal yields an effect of 0.043 versus 0.016 for matched random removal. Provenance discrimination is more modest by ranking AUROC (0.55-0.62), despite substantial top-k enrichment, revealing that provenance association and interventional faithfulness need not coincide. Our results establish synthetic provenance as a controlled setting for verifiable contributive attribution

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.02347 — 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 2026-10-05T23:01:38.567Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).