---
title: "More Agents, More Reliable? System-Level Reliability of Multi-Agent LLMs under Operational-Profile Uncertainty"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.07140v1"
permalink: "https://feed.astralor.com/items/itm_07f3d1a8ce8555af458e997b"
published: "2026-11-04T21:08:18.000Z"
score: 56
channel: "tech"
tags: ["LLM","AI Agent","RAG","More Agents","More Reliable","System-Level Reliability","Multi-Agent LLMs","Operational-Profile Uncertainty Teams"]
---

# More Agents, More Reliable? System-Level Reliability of Multi-Agent LLMs under Operational-Profile Uncertainty

## 速览

Teams increasingly build software from cooperating large language model (LLM) agents, often assuming more agents mean more reliability. Yet the usual evidence, average benchmark accuracy, ignores the input mix a deploye…

## 关键点

- Teams increasingly build software from cooperating large language model (LLM) agents, often assuming more agents mean more reliability.
- Yet the usual evidence, average benchmark accuracy, ignores the input mix a deployed system meets and the rule merging the agents' answers.
- Building on HIP-LLM, a hierarchical imprecise-probability assessment of single LLMs, we propose a framework combining it with a Bayesian network of the system's failure logic for multi-agent systems.

## 全文

Teams increasingly build software from cooperating large language model (LLM) agents, often assuming more agents mean more reliability. Yet the usual evidence, average benchmark accuracy, ignores the input mix a deployed system meets and the rule merging the agents' answers. Building on HIP-LLM, a hierarchical imprecise-probability assessment of single LLMs, we propose a framework combining it with a Bayesian network of the system's failure logic for multi-agent systems. Agent failure-probability intervals propagate through this network, where the combining rule can itself fail, into a reliability envelope showing which evidence limits it. In a pipeline extracting regulated chemical-hazard statements, majority voting over three models protected against the weakest but did not clearly exceed the best, and a voter failing on 5\% of demands would largely erase its advantage over the middle model. More agents yield a more defensible reliability claim, not automatically a more reliable system.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_07f3d1a8ce8555af458e997b