Metraipolis › archive › dc_e33746392bb8
Approval Integrity and Recovery in LLM Answer Publication — arXiv:2609.15576
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.15576
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T23:54:06.803Z
- SHA-256 of the archived text
658c19dd5414a88463b3a5c904e86ea5431b1f55a23b762adddbf78e5e9c53f4- Stable URL of this page
https://metraipolis.com/documento/dc_e33746392bb8
How to cite
Approval Integrity and Recovery in LLM Answer Publication — arXiv:2609.15576 — arXiv (en), https://arxiv.org/abs/2609.15576 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e33746392bb8 (Metraipolis archive dc_e33746392bb8, sha256 658c19dd5414a88463b3a5c904e86ea5431b1f55a23b762adddbf78e5e9c53f4)
BibTeX
@misc{dce33746392bb8,
title = {Approval Integrity and Recovery in LLM Answer Publication — arXiv:2609.15576},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.15576},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e33746392bb8; sha256 658c19dd5414a88463b3a5c904e86ea5431b1f55a23b762adddbf78e5e9c53f4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.15576. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Publication integrity in LLM systems requires binding approved content to its current authorization context. We examine exact-content binding, authorization freshness and checkpoint recovery in Lightcap's publication enforcement mechanism. On 900 independently human-annotated RAGTruth responses from 150 source tasks, three dated Ministral models and a same-model direct-grounding baseline yield 3,600 assessments. The production response-act checker instantiated with 14B accepts 291 of 302 unsupported-labelled answers; the direct baseline accepts 41. Supported-answer retention is 95.2% and 66.9%, respectively. An exact promotion-correction identity tracks error through 100 chronological 3B-14B-8B-14B answer trajectories. Among 65 initially approved answers, the final stateful recheck-recovery policy increases exact-match error by 9.23 percentage points relative to the initial checkpoint (95% article-clustered interval [-1.72, 19.61]). Controlled evidence-fingerprint changes expose asymmetric freshness enforcement between publication and recovery. A separate BIPIA prompt-injection experiment records zero target insertions among 266 valid editor outputs. External Hugging Face calibration experiments transfer retrieval models from ArguAna to SciFact and NFCorpus, and diagnostic decision rules from Thunderbird to BGL, distinguishing probability calibration from ranking changes. The measurements separate semantic false approval, stale authorization and recovery-induced error at executable publication boundaries. Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CR). Categorie arXiv: Cryptography and Security (cs.CR). Pagina del lavoro: https://arxiv.org/abs/2609.15576 — 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 2026-10-15T23:20:42.711Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).