TelcoLM: collecting data, adapting, and benchmarking language models for the telecommunication domain
This addresses the need for accurate LLMs in telecommunications, an industrial domain with many valuable use cases, but it is incremental as it applies existing adaptation methods to new data.
The paper tackled the problem of low accuracy of Large Language Models (LLMs) in the highly technical telecommunications domain by collecting a massive corpus of domain-specific data (800M tokens, 80K instructions) and adapting models, showing that domain-adapted models can challenge larger generalist models.
Despite outstanding processes in many tasks, Large Language Models (LLMs) still lack accuracy when dealing with highly technical domains. Especially, telecommunications (telco) is a particularly challenging domain due the large amount of lexical, semantic and conceptual peculiarities. Yet, this domain holds many valuable use cases, directly linked to industrial needs. Hence, this paper studies how LLMs can be adapted to the telco domain. It reports our effort to (i) collect a massive corpus of domain-specific data (800M tokens, 80K instructions), (ii) perform adaptation using various methodologies, and (iii) benchmark them against larger generalist models in downstream tasks that require extensive knowledge of telecommunications. Our experiments on Llama-2-7b show that domain-adapted models can challenge the large generalist models. They also suggest that adaptation can be restricted to a unique instruction-tuning step, dicarding the need for any fine-tuning on raw texts beforehand.