aiqbench analysis
This research addresses the critical bottleneck of data scarcity in network traffic classification by deploying lightweight Generative AI models that operate on compact flow-level representations rather than raw packet bytes. By utilizing transformer-based and state-space architectures with only a few million parameters, the study achieves high-fidelity synthetic data generation while maintaining computational efficiency suitable for practical deployment. The findings demonstrate that these streamlined approaches effectively preserve complex temporal traffic dynamics, enabling privacy-preserving model training and robust performance in low-data regimes.
Lightweight GenAI for Network Traffic Generation: Fidelity, Augmentation, and Classification
arXiv:2603.25507v2 Announce Type: replace-cross Abstract: Network Traffic Classification (NTC) increasingly relies on data-driven models, yet its practical deployment is often constrained by limited labeled data, strict privacy requirements, and the cost of collecting representative traffic traces. While Network Traffic Generation (NTG) provides an effective means to mitigate data scarcity, conventional generative methods struggle to model the complex temporal dynamics of modern traffic and often incur high computational costs. In this article, we investigate lightweight Generative Artificial Intelligence (GenAI) architectures for practical NTG. Rather than generating raw packet bytes or relying on large foundation models, we synthesize compact flow-level traffic representations derived from early packet-header information, enabling transformer-based, state-space, and diffusion models with only a few million parameters. We present a modular GenAI pipeline for NTG and evaluate it along four complementary axes: (i) synthetic traffic fidelity, (ii) synthetic-only training for privacy-preserving NTC, (iii) data augmentation under low-data regimes, and (iv) computational efficiency. Experiments on two heterogeneous datasets show that lightweight transformer-based and state-space models preserve both static and temporal traffic characteristics, while providing useful synthetic data for downstream NTC. Among them, transformer-based models offer the best fidelity-efficiency trade-off, combining high-quality traffic generation with moderate computational overhead.
Abstract:Network Traffic Classification (NTC) increasingly relies on data-driven models, yet its practical deployment is often constrained by limited labeled data, strict privacy requirements, and the cost of collecting representative traffic traces. While Network Traffic Generation (NTG) provides an effective means to mitigate data scarcity, conventional generative methods struggle to model the complex temporal dynamics of modern traffic and often incur high computational costs. In this article, we investigate lightweight Generative Artificial Intelligence (GenAI) architectures for practical NTG. Rather than generating raw packet bytes or relying on large foundation models, we synthesize compact flow-level traffic representations derived from early packet-header information, enabling transformer-based, state-space, and diffusion models with only a few million parameters. We present a modular GenAI pipeline for NTG and evaluate it along four complementary axes: (i) synthetic traffic fidelity, (ii) synthetic-only training for privacy-preserving NTC, (iii) data augmentation under low-data regimes, and (iv) computational efficiency. Experiments on two heterogeneous datasets show that lightweight transformer-based and state-space models preserve both static and temporal traffic characteristics, while providing useful synthetic data for downstream NTC. Among them, transformer-based models offer the best fidelity-efficiency trade-off, combining high-quality traffic generation with moderate computational overhead.
Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
C.2.3; C.2.4; C.2.5; I.2.1; I.2.m
arXiv:2603.25507 [cs.NI]
(or arXiv:2603.25507v2 [cs.NI] for this version)
https://doi.org/10.48550/arXiv.2603.25507 Focus to learn more arXiv-issued DOI via DataCite
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.
Full text under the publisher's open licence · 296 words
Related coverage
Digital States 2026: AI Moves From Experiment to Infrastructure
govtech.com · Sep 24, 2026
Article: Multi-Agent AI for Production Security Operations: an A2A and MCP Architecture in a 5G Core
InfoQ · Jul 23, 2026
How Ninth Wave built AI-powered open finance onboarding on Amazon Bedrock
AWS · Sep 14, 2026
A GAN-Based Framework for Robust DDoS Attack Detection
arXiv · Sep 17, 2026