Metraipolis › archive › dc_d4e746384737

Targeted Counterfactual Fingerprinting for Black-Box LLM Ownership Verification — arXiv:2608.08195

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.08195
License (as archived)
CC0-1.0
Archived
2026-10-19T08:58:13.995Z
SHA-256 of the archived text
58cd7a81da433d4fe829163449a91979db8aabeab3dff9bda2fb04f7a6fc490c
Stable URL of this page
https://metraipolis.com/documento/dc_d4e746384737

How to cite

Targeted Counterfactual Fingerprinting for Black-Box LLM Ownership Verification — arXiv:2608.08195 — arXiv (en), https://arxiv.org/abs/2608.08195 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d4e746384737 (Metraipolis archive dc_d4e746384737, sha256 58cd7a81da433d4fe829163449a91979db8aabeab3dff9bda2fb04f7a6fc490c)
BibTeX
@misc{dcd4e746384737,
  title = {Targeted Counterfactual Fingerprinting for Black-Box LLM Ownership Verification — arXiv:2608.08195},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.08195},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d4e746384737; sha256 58cd7a81da433d4fe829163449a91979db8aabeab3dff9bda2fb04f7a6fc490c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.08195. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) are high-value assets that can be derived through redeployment, fine-tuning, quantization, or further alignment. Because deployed LLMs are commonly exposed only through query APIs, ownership verification must often rely on black-box text responses. This setting is difficult: generations are open-ended and can vary across repeated queries, while existing black-box fingerprints rely on signals that are fragile under a final-response interface, including full-text matching, soft behavioral features, or model-specific prompts designed not to transfer. We propose TCF (Targeted Counterfactual Fingerprinting), a black-box LLM fingerprinting framework that converts open-ended generation comparison into constrained-answer targeted counterfactual transfer. TCF restricts each verification query to a finite answer space, reducing the surface-form ambiguity that enters the verification score, and optimizes a prompt perturbation toward a counterfactual target different from the protected model's clean answer on the original prompt. Verification reduces to checking whether the suspect model's parsed final answer matches the recorded target. We introduce the source-model counterfactual margin (SCM), a protected-model-only quantity that certifies the target is unlikely before the perturbation and likely after it; SCM controls target selection, perturbation stopping, and fingerprint filtering. Under explicit derived-preservation and independent-transfer budgets motivated by local behavioral closeness, we derive a target-accuracy gap between derived and independent models. Across four LLM families, TCF achieves an average AUC of 0.9861, improving over TRAP, ProFLingo, and ZeroPrint by 0.07 to 0.19.

Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-08-11 in cs.CR).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.08195 — 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 2026-10-19T08:54:24.938Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-08-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).