提出量化大模型多智能体输出耦合的工具，揭示视觉条件可能夸大性能表现。
The Contagion Tensor: A Framework for Measuring Output-Distribution Coupling in Multi-Agent LLM Systems -- and Auditing the Claims It Enables
- 构建传播张量框架，通过基准比值度量跨模态、智能体和时间的输出耦合强度。
- 实验证明视觉条件下的超线性效应（CAF=1.40）在关闭图像扰动后降为亚线性（CAF=0.87）。
- 适用于评估多智能体系统中真实耦合与设计伪影，适合模型审计与仿真验证。
我们提出传播张量（Contagion Tensor），用于量化大语言模型（LLM）在模态、智能体和时间步之间的输出分布耦合程度。从中推导出耦合放大因子（CAF），一种基于比率的无量纲指标，形式为 CAF = E[T_condition] / E[T_baseline]，并提供自助法置信区间。我们实现四种CAF变体，并在完整的2x2x2块正交模拟设计中评估最强版本，进行模态特异性消融。消融显示，看似图像条件下的超线性效应（CAF=1.40）在关闭图像扰动模块后降至亚线性（CAF=0.87），变化-0.53，而文本条件影响不变。补充的实时API实验涵盖两个模型族：DeepSeek-Chat（R=30）和GPT-4o-mini（R=15，真实视觉）。在统一角色下，纯文本通信的CAF≈1.0；多样角色引发收敛（CAF=0.88）。对GPT-4o-mini的模型内比较显示：C3（文本）CAF=1.02 vs. C5（真实视觉，R=30）CAF=1.72 [1.700, 1.733]，增量+0.70，验证了模拟中的超线性预测。11个条件中已有5个在真实API上测试，6个尚未验证。贡献在于：（1）一个使输出耦合可量化解构的测量工具；（2）一种可迁移的消融协议，任何模块化多智能体模拟器均可采用以区分真实耦合与设计伪影。
原文摘要 · Abstract (English)
We introduce the Contagion Tensor, a measurement framework for quantifying how large language model (LLM) output distributions couple across modalities, agents, and time steps. From the tensor we derive the Coupling Amplification Factor (CAF), a family of ratio-based metrics sharing the form CAF = E[T_condition] / E[T_baseline], providing unitless, baseline-referenced measurement with bootstrap confidence intervals. We instantiate CAF in four variants and evaluate the strongest in a complete 2x2x2 block-orthogonal simulation design with modality-specific ablation. The ablation reveals that an apparent image-condition super-linear effect (CAF = 1.40) collapses to sub-linear (CAF = 0.87) when the image perturbation module is disabled, a shift of -0.53 with zero effect on text conditions. We supplement with real-API experiments across two model families: DeepSeek-Chat (R=30) and GPT-4o-mini (R=15, real vision). Under uniform personas, text-only communication produces CAF approx 1.0 in both models. Diverse personas drive convergence (CAF = 0.88). A within-model comparison on GPT-4o-mini reveals: C3 (text) CAF = 1.02 vs. C5 (real vision, R=30) CAF = 1.72 [1.700, 1.733], delta = +0.70, validating the simulation's super-linear image-condition prediction. Of 11 conditions, 5 have been tested on real APIs and 6 remain unverified. Our contribution is two-layered: (1) a measurement instrument that makes output-distribution coupling quantitatively falsifiable; and (2) a transferable ablation protocol that any modular multi-agent simulator can adopt to distinguish genuine coupling from design artifacts.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。