---
title: "Fidelity Collapses with Table Size, Not Turn Count: Delivery Architecture in a Deployed Tool-Using Agent"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.00442v1"
permalink: "https://feed.astralor.com/items/itm_def919a3306d54888576e525"
published: "2026-10-30T22:50:32.000Z"
score: 55
channel: "ai"
tags: ["AI Agent","OpenAI","Google","Fidelity Collapses","Table Size","Not Turn Count","Delivery Architecture","Deployed Tool-Using Agent"]
---

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

## At a glance

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 ag…

## Key points

- 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, a…

## Full text

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.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_def919a3306d54888576e525