Metraipolis › archive › dc_2dca5fba5eb5

Generation Provenance Before Behavior Attribution: Auditing Synthetic Speech Research Objects — arXiv:2610.01378

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.01378
License (as archived)
CC0-1.0
Archived
2026-10-06T12:04:49.983Z
SHA-256 of the archived text
04792482b315e182bf25ce2c39538ec9a4d129f57df1a8f5609554fe68c13348
Stable URL of this page
https://metraipolis.com/documento/dc_2dca5fba5eb5

How to cite

Generation Provenance Before Behavior Attribution: Auditing Synthetic Speech Research Objects — arXiv:2610.01378 — arXiv (en), https://arxiv.org/abs/2610.01378 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2dca5fba5eb5 (Metraipolis archive dc_2dca5fba5eb5, sha256 04792482b315e182bf25ce2c39538ec9a4d129f57df1a8f5609554fe68c13348)
BibTeX
@misc{dc2dca5fba5eb5,
  title = {Generation Provenance Before Behavior Attribution: Auditing Synthetic Speech Research Objects — arXiv:2610.01378},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.01378},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2dca5fba5eb5; sha256 04792482b315e182bf25ce2c39538ec9a4d129f57df1a8f5609554fe68c13348},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01378. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Attributing model behavior to synthetic training data requires knowing what produced each training item before estimating what that item caused. A waveform-label pair does not preserve this knowledge. We propose a generation-provenance substrate in which a synthetic research object binds source specification, generated content, waveform, target, fact requirements, quality signals, review lineage, and immutable manifest identity. Producer and selection mechanism determine evidentiary meaning; storage location and variable name do not. We audit this substrate in a private Japanese care-handoff pipeline. A 113-asset review population contains 1.552 hours of synthetic speech across six scenario families; all items have linked audio, transcripts, candidate notes, and fact checklists, but human evidence is selective and source-specific. Two faithful-only manifests are scenario-seed-disjoint and immutably versioned, while exact upstream attribution remains blocked by floating generator aliases, missing per-clip TTS and code stamps, and an unversioned checking prompt. We argue that generation provenance is necessary but not sufficient for behavior attribution: it defines the candidate causal graph and audit units, whereas contributive attribution still requires frozen training runs and intervention or influence evidence. The paper contributes a compact provenance contract, an audit protocol, and a bounded case study for synthetic-data attribution; controlled research access may be offered, but we do not claim causal training-data attribution, clinical validity, or unrestricted public release.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.01378 — 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 2026-10-06T10:21:27.942Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).