Talk Like a Packet: Rethinking Network Traffic Analysis with Transformer Foundation Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917254538461184 |
|---|---|
| author | Mayhoub, Samara Foh, Chuan Heng Mashhadi, Mahdi Boloursaz Shojafar, Mohammad Tafazolli, Rahim |
| author_facet | Mayhoub, Samara Foh, Chuan Heng Mashhadi, Mahdi Boloursaz Shojafar, Mohammad Tafazolli, Rahim |
| contents | Inspired by the success of Transformer-based models in natural language processing, this paper investigates their potential as foundation models for network traffic analysis. We propose a unified pre-training and fine-tuning pipeline for traffic foundation models. Through fine-tuning, we demonstrate the generalizability of the traffic foundation models in various downstream tasks, including traffic classification, traffic characteristic prediction, and traffic generation. We also compare against non-foundation baselines, demonstrating that the foundation-model backbones achieve improved performance. Moreover, we categorize existing models based on their architecture, input modality, and pre-training strategy. Our findings show that these models can effectively learn traffic representations and perform well with limited labeled datasets, highlighting their potential in future intelligent network analysis systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_06636 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Talk Like a Packet: Rethinking Network Traffic Analysis with Transformer Foundation Models Mayhoub, Samara Foh, Chuan Heng Mashhadi, Mahdi Boloursaz Shojafar, Mohammad Tafazolli, Rahim Networking and Internet Architecture Inspired by the success of Transformer-based models in natural language processing, this paper investigates their potential as foundation models for network traffic analysis. We propose a unified pre-training and fine-tuning pipeline for traffic foundation models. Through fine-tuning, we demonstrate the generalizability of the traffic foundation models in various downstream tasks, including traffic classification, traffic characteristic prediction, and traffic generation. We also compare against non-foundation baselines, demonstrating that the foundation-model backbones achieve improved performance. Moreover, we categorize existing models based on their architecture, input modality, and pre-training strategy. Our findings show that these models can effectively learn traffic representations and perform well with limited labeled datasets, highlighting their potential in future intelligent network analysis systems. |
| title | Talk Like a Packet: Rethinking Network Traffic Analysis with Transformer Foundation Models |
| topic | Networking and Internet Architecture |
| url | https://arxiv.org/abs/2602.06636 |