Metraipolis › archive › dc_687b015610fa

Verification as an Architectural Layer for LLM Agents: A V-Model Design, and a Pilot Study of Its Deterministic Core — arXiv:2609.31937

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.31937
License (as archived)
CC0-1.0
Archived
2026-10-07T20:59:42.400Z
SHA-256 of the archived text
91dac34814d99ebdb5ff1553320a3553deaab7bd62d7264de94e4a8d3356436a
Stable URL of this page
https://metraipolis.com/documento/dc_687b015610fa

How to cite

Verification as an Architectural Layer for LLM Agents: A V-Model Design, and a Pilot Study of Its Deterministic Core — arXiv:2609.31937 — arXiv (en), https://arxiv.org/abs/2609.31937 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_687b015610fa (Metraipolis archive dc_687b015610fa, sha256 91dac34814d99ebdb5ff1553320a3553deaab7bd62d7264de94e4a8d3356436a)
BibTeX
@misc{dc687b015610fa,
  title = {Verification as an Architectural Layer for LLM Agents: A V-Model Design, and a Pilot Study of Its Deterministic Core — arXiv:2609.31937},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.31937},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_687b015610fa; sha256 91dac34814d99ebdb5ff1553320a3553deaab7bd62d7264de94e4a8d3356436a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31937. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language model (LLM) agents built on the ReAct pattern concentrate four responsibilities in one model: selecting a strategy, choosing each action, formatting it, and judging whether the result is adequate. Nothing outside the generative loop can reject its output, so an agent that cannot make progress does not report failure; it runs until an external budget stops it. We propose treating verification as an architectural layer by adapting the V-model from software engineering: specification levels descend from requirements to individual steps, each level is paired with a dedicated verifier, a deterministic controller enforces every verdict, and only verification outcomes write to memory, so a rejection localizes the level that introduced the fault and an agent halts by declining rather than by exhaustion. Each verifier separates a zero-cost deterministic \emph{gate} from an optional LLM \emph{judge}, so the contribution and cost of each can be measured independently. We report a pilot implementing the acceptance- and unit-level verifier pairs, comparing five configurations that share one executor, tool set, and scorer and differ only in verification, on the four-hop stratum of MuSiQue with an 8B-parameter backbone. Across 47 executions, the two unverified configurations answered none of ten questions, every run ending at a step cap or provider token limit; the verified configuration without a planner answered eight and abstained on the rest. Deterministic gates produced eight of the nine observed corrections at zero marginal cost, and planning degraded performance once verification was present. These results characterize termination behavior, not accuracy at scale; we outline a twelve-month plan to complete and evaluate the full architecture, including the integration-level pair the pilot omits.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI).
Categorie arXiv: Software Engineering (cs.SE) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.31937 — 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 2026-10-07T20:29:02.185Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).