Metraipolis › archive › dc_9e37cd730f31
A Benchmark and Diagnostic Study of Epistemic Admission in Shared Agent Memory — arXiv:2609.30813
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.30813
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T01:38:21.803Z
- SHA-256 of the archived text
f88a4186925123e23116a6824b84dd55a9e806179295bdaa3c50b28df30a8f94- Stable URL of this page
https://metraipolis.com/documento/dc_9e37cd730f31
How to cite
A Benchmark and Diagnostic Study of Epistemic Admission in Shared Agent Memory — arXiv:2609.30813 — arXiv (en), https://arxiv.org/abs/2609.30813 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9e37cd730f31 (Metraipolis archive dc_9e37cd730f31, sha256 f88a4186925123e23116a6824b84dd55a9e806179295bdaa3c50b28df30a8f94)
BibTeX
@misc{dc9e37cd730f31,
title = {A Benchmark and Diagnostic Study of Epistemic Admission in Shared Agent Memory — arXiv:2609.30813},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.30813},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9e37cd730f31; sha256 f88a4186925123e23116a6824b84dd55a9e806179295bdaa3c50b28df30a8f94},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30813. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Evaluating claim admission in shared agent memory is challenging because repeated claims may be mistaken for independent evidence. An agent may copy or paraphrase a retrieved belief, while admitting a false claim exposes subsequent agents to it. To study this problem, we introduce the Correlated Promotion Benchmark (CPB), which evaluates whether candidate claims should be admitted to shared this http URL -Static constructs a frozen test split from publicly annotated sources with fixed gold actions. CPB-Live runs multi-agent teams over a shared store, records all writes and retrievals, and tracks source lineage defined by each scenario. A separate consumer answers from the store alone. We evaluate eight admission policies across four agent families. Our results show that policies which deduplicate sources reject many true claims alongside false ones, whereas policies preserving answer coverage admit nearly as many false claims as unrestricted sharing. Gating on declared source type reduces false adoption to 0.06--0.09, compared with 0.22--0.47 for other answering policies. Once an uncontested false belief enters memory, the consumer asserts it in 0.97--0.99 of probes across all families. No non-oracle policy consistently rejects false claims across verbatim copies, paraphrases, and paraphrases declared authoritative. These findings reveal the limitations of admission policies without access to source lineage. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-28 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.30813 — 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 2026-10-08T01:12:14.293Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).