LLMs for Networking

A Large Language Model (LLM) is a type of artificial intelligence trained on massive amounts of text data to understand, generate, and manipulate human language. By analyzing patterns, context, and relationships between words, LLMs can perform a wide range of task such as answering questions, writing content, summarizing text, translating languages, and even generating code based on a given prompt or instruction.

Large Language Models (LLMs) are increasingly being applied to the challenges of navigating and implementing Open RAN (O-RAN) specifications, primarily to help engineers manage the immense complexity of technical documents from bodies like 3GPP and the O-RAN Alliance. To ensure accuracy and reduce model hallucinations, most applications rely on Retrieval-Augmented Generation (RAG) to ground responses in verifiable source documents, or involve fine-tuning open-source models on specialized O-RAN datasets. This enables a wide range of practical applications, including standards comprehension, automated test case generation, intelligent network control, and AI-assisted code development.

Research Scholars

Yaswanth Kumar L. S., Michael Suguna Kumar Victor, Abdulla Ovais