Metraipolis › archive › dc_6823bffee891

Stale Does Not Mean Unsafe: Guard Precision for Tool-Using LLM Agents under Infrastructure State Races — arXiv:2609.29522

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.29522
License (as archived)
CC0-1.0
Archived
2026-10-08T04:26:12.757Z
SHA-256 of the archived text
abd2d6a47ce797d887c716633399f5779a4d12c07ae8837d0048810070d643e5
Stable URL of this page
https://metraipolis.com/documento/dc_6823bffee891

How to cite

Stale Does Not Mean Unsafe: Guard Precision for Tool-Using LLM Agents under Infrastructure State Races — arXiv:2609.29522 — arXiv (en), https://arxiv.org/abs/2609.29522 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6823bffee891 (Metraipolis archive dc_6823bffee891, sha256 abd2d6a47ce797d887c716633399f5779a4d12c07ae8837d0048810070d643e5)
BibTeX
@misc{dc6823bffee891,
  title = {Stale Does Not Mean Unsafe: Guard Precision for Tool-Using LLM Agents under Infrastructure State Races — arXiv:2609.29522},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.29522},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6823bffee891; sha256 abd2d6a47ce797d887c716633399f5779a4d12c07ae8837d0048810070d643e5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29522. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Tool-using language-model agents increasingly mutate schedulers, data pipelines, object stores, and access-control systems. Between an agent's read and its commit, external state can change, but not every change makes the commit unsafe. We separate invalidating races, which break a declared safety predicate, from predicate-preserving and irrelevant races, and ask how precisely runtime guards distinguish them. Our deterministic simulator separates visible from authoritative state and injects five non-atomic failure mechanisms across 16 infrastructure tasks in four domains; frozen agent proposals are replayed counterfactually under every controller without an LLM judge. We evaluate three commit-time guard granularities (global epoch, read-set version, semantic commit predicate), multi-level verification, and model-side gates on three locally hosted quantized model families (Qwen3-4B, Phi-4-mini, Gemma4-8B; 3,456 trajectories on one GPU). All three guards eliminate unsafe commits, but their availability differs sharply: freshness-based guards needlessly block 92-95% of benign races, forfeiting up to 43% of safe task completions, while the complete predicate guard blocks none. That precision is contract-dependent: deleting a single declared clause converts exactly its fault family into unsafe commits (up to 7.9%). Model-side signals do not substitute: verbal confidence is miscalibrated (ECE approximately 0.37), action agreement matches a random gate, a cautionary prompt leaves the direct unsafe rate essentially unchanged, and after a freshness-guard block agents re-commit unsafely from refreshed but still-incomplete reads. Under degraded telemetry a hidden concurrent mutation remains observationally clean, bounding every selective policy. Precise runtime enforcement therefore requires semantic contracts, not freshness heuristics or model self-assessment.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.29522 — 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 2026-10-08T03:50:57.783Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).