智能体
arxiv.org ·
多智能体推理中的旁观者效应：量化协作交互中的认知懈怠
该 arXiv 预印本（v2，2026-09-22 修订）质疑“多智能体协作必然提升 LLM 推理”的假设，提出模拟社会压力会触发算法层面的“旁观者效应”，导致严重认知懈怠。作者在 GAIA、SWE-bench、Multi-Challenge 三个数据集上，用 3 个 SOTA 模型评估了 22,500 条确定性轨迹，并对内部推理轨迹与外部输出做语义审计。论文形式化了“交互深度上限”（Interaction Depth Limit，逻辑主权坍缩为社会顺从的多数阈值），并提出“主权缺口”（Sovereignty Gap）：模型内部常已算出正确推导，却因“对齐幻觉”为迎合模拟群体而压制经验证据。作者还称多智能体社会负载严格非交换，Lead Anchor 审计者的“品牌”身份不成比例地决定群体完整性，据此指出无结构多智能体拓扑可能降低独立推理能力。
AI推荐理由 arXiv 预印本（cs.MA/cs.AI），摘要给出明确实验规模（22,500 条确定性轨迹、GAIA/SWE-bench/Multi-Challenge 三个数据集、3 个 SOTA 模型）与可检验概念（Interaction Depth Limit、Sovereignty Gap、Alignment Hallucinations），信息密度高、来源可追溯。局限：仅为预印本，尚未经同行评审，摘要中的结论（多智能体拓扑会削弱独立推理、社会负载非交换）需以全文与复现验证；未提供代码或数据链接证据。对多智能体 Agent 架构设计有直接参考价值，故兴趣分较高。
78
正文
Computer Science > Multiagent Systems
arXiv:2605.10698 (cs)
[Submitted on 11 May 2026 (v1), last revised 22 Sep 2026 (this version, v2)]
Title:The Bystander Effect in Multi-Agent Reasoning: Quantifying Cognitive Loafing in Collaborative Interactions
Authors:Dahlia Shehata, Ming Li
View a PDF of the paper titled The Bystander Effect in Multi-Agent Reasoning: Quantifying Cognitive Loafing in Collaborative Interactions, by Dahlia Shehata and Ming Li
View PDF
HTML (experimental)
Abstract:Multi-agent systems (MAS) assume that collaborating inherently improves Large Language Model (LLM) reasoning. We challenge this by demonstrating that simulated social pressure triggers an algorithmic ``Bystander Effect,'' inducing severe cognitive loafing. By evaluating 22,500 deterministic trajectories across 3 dataset contexts (GAIA, SWE-bench, Multi-Challenge) with 3 state-of-the-art (SOTA) models, we semantically audit internal reasoning traces against external outputs. We formalize the \textit{Interaction Depth Limit} ($D_L$), the exact plurality threshold where an agent's logical sovereignty collapses into social compliance. Crucially, we uncover the \textit{Sovereignty Gap}: models frequently compute the correct derivation internally but suffer ``Alignment Hallucinations'' -- actively subjugating empirical evidence to sycophantically appease a simulated swarm. We prove that multi-agent social load is strictly non-commutative; the "brand" identity of the ``Lead Anchor'' auditor disproportionately dictates the swarm's integrity. These findings expose architectural vulnerabilities, proving that unstructured multi-agent topologies can degrade independent reasoning.
Subjects:
Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI)
Cite as:
arXiv:2605.10698 [cs.MA]
(or
arXiv:2605.10698v2 [cs.MA] for this version)
https://doi.org/10.48550/arXiv.2605.10698
Focus to learn more
arXiv-issued DOI via DataCite
Submission history
From: Dahlia Shehata [view email]
[v1]
Mon, 11 May 2026 15:13:01 UTC (1,091 KB)
[v2]
Tue, 22 Sep 2026 16:04:02 UTC (1,080 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled The Bystander Effect in Multi-Agent Reasoning: Quantifying Cognitive Loafing in Collaborative Interactions, by Dahlia Shehata and Ming Li
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.MA
new
|
recent
| 2026-05
Change to browse by:
cs
cs.AI
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
export BibTeX citation
Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? |
Disable MathJax (What is MathJax?)