An Empirical Study of NetOps Capability of Pre-Trained Large Language Models
This work addresses the need for domain-specific evaluation in networks for industry practitioners, but it is incremental as it applies existing methods to a new dataset.
The authors introduced NetEval, a 5,732-question evaluation set for assessing the Network Operations (NetOps) capabilities of large language models across five sub-domains, and found that only GPT-4 performs competitively with humans, while open models like LLaMA 2 show potential.
Nowadays, the versatile capabilities of Pre-trained Large Language Models (LLMs) have attracted much attention from the industry. However, some vertical domains are more interested in the in-domain capabilities of LLMs. For the Networks domain, we present NetEval, an evaluation set for measuring the comprehensive capabilities of LLMs in Network Operations (NetOps). NetEval is designed for evaluating the commonsense knowledge and inference ability in NetOps in a multi-lingual context. NetEval consists of 5,732 questions about NetOps, covering five different sub-domains of NetOps. With NetEval, we systematically evaluate the NetOps capability of 26 publicly available LLMs. The results show that only GPT-4 can achieve a performance competitive to humans. However, some open models like LLaMA 2 demonstrate significant potential.