Metraipolis › archive › dc_ffbb95494651

Fidelity Collapses with Table Size, Not Turn Count: Delivery Architecture in a Deployed Tool-Using Agent — arXiv:2611.00442

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.00442
License (as archived): CC0-1.0
Archived: 2026-11-03T10:34:12.996Z
SHA-256 of the archived text: 5b962e1d4003606ec08283c05405e037e3503b1ecde3ce1d60717d46e4019a64
Stable URL of this page: https://metraipolis.com/documento/dc_ffbb95494651

How to cite

Fidelity Collapses with Table Size, Not Turn Count: Delivery Architecture in a Deployed Tool-Using Agent — arXiv:2611.00442 — arXiv (en), https://arxiv.org/abs/2611.00442 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ffbb95494651 (Metraipolis archive dc_ffbb95494651, sha256 5b962e1d4003606ec08283c05405e037e3503b1ecde3ce1d60717d46e4019a64)

BibTeX

@misc{dcffbb95494651,
  title = {Fidelity Collapses with Table Size, Not Turn Count: Delivery Architecture in a Deployed Tool-Using Agent — arXiv:2611.00442},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.00442},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ffbb95494651; sha256 5b962e1d4003606ec08283c05405e037e3503b1ecde3ce1d60717d46e4019a64},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00442. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

When a multi-turn agent delivers a tool's structured records into the model's context and re-emits them each turn, every turn can corrupt data the tool computed correctly. We measure this in a production award-travel agent: two architectures over identical frozen payloads, lineage-matched prompts and sampling, differing only in delivery architecture. SYNTHESIS feeds full record sets into the planner's context and answers follow-ups in-head; BYPASS, the deployed system, keeps records in session state, translates follow-ups into arguments for a deterministic filter, and delivers rows over a typed side channel. Across 24 dialogues (339 scored turns per condition), BYPASS scores 92.3% of turns fully correct, SYNTHESIS 45.7% (paired gap +46.6pp [+35.7, +57.1]). Varying delivered-table size by capture date, SYNTHESIS collapses from near-perfect at 10 rows or fewer to 8% or less beyond about 74 rows. On a content-controlled prefix ladder the collapse threshold orders by model capability, from under 10 rows (GPT-4o-mini) to about 198 or more (Gemini 2.5 Pro), while BYPASS stays size-flat for all four models. The collapse is expensive: beyond 100 rows, where almost no SYNTHESIS answer is correct, each correct one costs roughly 200x BYPASS's flat \$0.012. Practitioners converged on typed side-channel delivery for rendering, not correctness; we show it was, unknowingly, also a correctness decision.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.00442 — 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 2026-11-03T10:29:48.695Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).