Metraipolis › archive › dc_2a8d2c293d1b

When Do PEFT Adaptations Leak Structure? Measuring Black-Box Structural Bounds in Public-Base Model Services — arXiv:2608.05036

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.05036
License (as archived)
CC0-1.0
Archived
2026-10-31T11:27:33.672Z
SHA-256 of the archived text
3617ced037a48cba38caf41aea63d3193753f14a5c89fdae4244d733641df24e
Stable URL of this page
https://metraipolis.com/documento/dc_2a8d2c293d1b

How to cite

When Do PEFT Adaptations Leak Structure? Measuring Black-Box Structural Bounds in Public-Base Model Services — arXiv:2608.05036 — arXiv (en), https://arxiv.org/abs/2608.05036 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2a8d2c293d1b (Metraipolis archive dc_2a8d2c293d1b, sha256 3617ced037a48cba38caf41aea63d3193753f14a5c89fdae4244d733641df24e)
BibTeX
@misc{dc2a8d2c293d1b,
  title = {When Do PEFT Adaptations Leak Structure? Measuring Black-Box Structural Bounds in Public-Base Model Services — arXiv:2608.05036},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.05036},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2a8d2c293d1b; sha256 3617ced037a48cba38caf41aea63d3193753f14a5c89fdae4244d733641df24e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05036. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Services increasingly deploy public foundation models with private parameter-efficient adaptations, creating a differential information leakage risk when auditors or adversaries can execute the public base model locally and observe victim outputs. We present VectorHijack-SR, a measurement methodology that converts paired victim/base residuals into calibrated structural bounds over PEFT family, layer locality, and coarse rank, while separating metadata visibility from open-world validity and operational exploitability. Our estimator aggregates query-level magnitude, ranking, entropy, margin, length, template, locality, and spectral statistics into service-level representations. A service-disjoint classifier quantifies structural evidence, and a cross-fitted hierarchical rejector evaluates whether a victim lies outside the calibrated LoRA manifold. Across classification backbones, family leakage exceeds uniform chance on BERT/MNLI (8/12), RoBERTa/MNLI (21/24), and DeBERTa-v3 on MNLI (12/18) and AG News (15/18). Rank inference is task dependent: BERT/MNLI and DeBERTa/AG News reach 8/9, whereas DeBERTa/MNLI achieves 4/9 and is statistically compatible with chance after correction. On a ten-seed BERT open-set grid, the rejector achieves pooled AUROC 0.804 (95% CI [0.660, 0.927]) and known accuracy 0.956, but remains limited on structurally close DoRA and LoRA+head variants. Exact-version linkage on five held-out LoRA-r64 services reaches AUC 0.940. Experiments reveal a visibility-exploitability gap: two-stage recovery provides no fair-budget query savings, posterior-selected PEFT underperforms distill-then-convert PEFT (0.356 vs. 0.517), and free-running generation remains near chance. These results show that known-base, rich-output PEFT services can leak actionable structural and version information, while closed-set confidence alone does not establish universal adapter recovery.

Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-08-06 in cs.CR).
Categorie arXiv: Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2608.05036 — 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 2026-10-31T11:17:19.453Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-08-06?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).