剖析网络基础模型的隐含知识，发现其表征存在显著缺陷。
Demystifying Network Foundation Models
- 通过几何分析、度量对齐与因果敏感性测试，系统评估模型表征质量。
- 所有模型均存在各向异性、特征敏感性不一致、上下文分离能力差等问题。
- 揭示共性缺陷，优化后性能可提升0.35分，无需修改模型结构。
本文系统研究了网络基础模型（NFMs）中编码的隐含知识，重点聚焦于隐藏表示的分析而非下游任务表现。不同于现有工作，我们采用三部分评估：嵌入几何分析以评估表征空间利用效率，度量对齐评估以衡量与领域专家特征的一致性，因果敏感性测试以评估对协议扰动的鲁棒性。基于涵盖受控环境与真实场景的五个多样化网络数据集，我们评估了四种最先进的NFMs，发现它们普遍存在显著各向异性、特征敏感性模式不一致、无法有效分离高层上下文、负载依赖性及其他属性。本研究识别出所有模型中的多项局限，并证明通过针对性改进可显著提升性能（不改变架构时，F₁分数最高提升0.35）。
原文摘要 · Abstract (English)
This work presents a systematic investigation into the latent knowledge encoded within Network Foundation Models (NFMs) that focuses on hidden representations analysis rather than pure downstream task performance. Different from existing efforts, we analyze the models through a three-part evaluation: Embedding Geometry Analysis to assess representation space utilization, Metric Alignment Assessment to measure correspondence with domain-expert features, and Causal Sensitivity Testing to evaluate robustness to protocol perturbations. Using five diverse network datasets spanning controlled and real-world environments, we evaluate four state-of-the-art NFMs, revealing that they all exhibit significant anisotropy, inconsistent feature sensitivity patterns, an inability to separate the high-level context, payload dependency, and other properties. Our work identifies numerous limitations across all models and demonstrates that addressing them can significantly improve model performance (by up to +0.35 $F_1$ score without architectural changes).
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。