EXECUTIVE SUMMARY

The integration of Generative AI and Large Language Models (LLMs) into network security and operational management presents transformative opportunities for enhancing application identification and traffic classification. Traditional methods that rely on literal matching and regex-based software flows face limitations in handling complex network tasks, particularly in deep packet inspection (DPI) and encrypted traffic analysis. This research proposes a hybrid architecture that leverages multiple LLM backends, such as GPT-2 and ModernBERT, to facilitate dynamic, context-aware, and adaptive traffic classification systems. While earlier works like TrafficGPT demonstrated the potential of LLMs in encrypted traffic classification, our research advances these capabilities by integrating batch processing and optimized edge inference techniques—directly addressing the scalability and performance constraints of prior approaches. We conducted comprehensive evaluations, including workload characterization and hardware-specific optimizations on Intel® Xeon® processors and Intel® Arc™ A770 Graphics. These efforts establish a practical foundation for deploying LLM-based systems at scale, extending the frontier of traditional network security.

Intel Labs is excited to present six works at this year’s ICML conference in Vancouver Canada, including two spotlight papers, one of which is also an oral presentation. Intel researchers will also present two works at related conference workshops

Souvik Kundu is a Staff Research Scientist at Intel Labs, leading scalable and efficient AI research to make AI more sustainable, efficient, and deployable across various hardware platforms. Souvik recently received the prestigious IEEE/ACM DAC Under-40 Innovators Award 2025 for his works on foundation model efficiency for resource limited hardware

Wondering if your AI startup should team up with a big tech company? Here’s what 350+ founders learned from working directly with Intel and how it paid off.

Utilizing block pruning techniques, Intel Labs researchers developed the Mamba-Shedder solution to remove redundancies in Mamba-based models, improving their computational and memory efficiency

Industrial enterprises need more than just AI to scale technology across diverse environments. They need open, flexible, and interoperable platforms.

Generative AI is changing how businesses work, and Retrieval-Augmented Generation (RAG) is one of the most exciting tools out there. It helps AI give smarter, more accurate answers by connecting it to your company’s data. But setting it up can be tricky—until now.

If you’re curious about how to bring AI into your business without the headaches, this post is for you.

Vector search is at the core of the AI revolution, and this blog series is here to teach you all about it. Our blog series introduces scalable vector search and dives deeper into sector compression, dimensionality reduction, and Retrieval-Augmented Generation systems.

Agentic AI is here: See how Intel® Liftoff startups are building smarter, more autonomous systems that plan, reason, and execute across real-world workflows.