Metraipolis › archive › dc_82f6dbdac81f
DP-MemView: A Memory Interface for Attribute-Level Transcript Privacy in Long-Term LLM Agents — arXiv:2608.03130
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.03130
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T12:31:24.147Z
- SHA-256 of the archived text
95b1521d0d69fae3bbd202d61b629c1e929a02b289912667e2a7e62c9f17e1ea- Stable URL of this page
https://metraipolis.com/documento/dc_82f6dbdac81f
How to cite
DP-MemView: A Memory Interface for Attribute-Level Transcript Privacy in Long-Term LLM Agents — arXiv:2608.03130 — arXiv (en), https://arxiv.org/abs/2608.03130 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_82f6dbdac81f (Metraipolis archive dc_82f6dbdac81f, sha256 95b1521d0d69fae3bbd202d61b629c1e929a02b289912667e2a7e62c9f17e1ea)
BibTeX
@misc{dc82f6dbdac81f,
title = {DP-MemView: A Memory Interface for Attribute-Level Transcript Privacy in Long-Term LLM Agents — arXiv:2608.03130},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.03130},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_82f6dbdac81f; sha256 95b1521d0d69fae3bbd202d61b629c1e929a02b289912667e2a7e62c9f17e1ea},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.03130. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Long-term memory enables persistent personalization in LLM agents, but repeated memory-conditioned responses can cumulatively reveal protected attributes even when they are never stated explicitly. We formalize this threat as adaptive transcript privacy and introduce DP-MemView, a differentially private interface that privately selects public response-conditioning views and exposes those views---rather than raw memory---to the response LLM. Each private selection is charged to every protected attribute whose memory group intersects the read set. Per-attribute ledgers block any selection that would exceed its cap and return a fixed generic view instead. Under an explicit interface contract, we prove pure B_a-DP for the entire adaptive transcript. We also extend the result to stores that differ across multiple protected groups and bound how much observing the transcript can change an adversary's prior odds. We evaluate the online and preallocated modes with three response LLMs on a controlled adjacent-store benchmark and a public-corpus transfer track. Both modes keep transcript distinguishability near chance while preserving target-required personalization and overall response quality. Further diagnostics show that removing key safeguards causes mismatched output support, missing ledger charges, revealing side channels, or growing long-horizon leakage. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.CL). Categorie arXiv: Cryptography and Security (cs.CR) ; Computation and Language (cs.CL); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.03130 — 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 2026-10-31T12:17:37.808Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).