English

Teaching Specific Scientific Knowledge into Large Language Models through Additional Training

Computation and Language 2023-12-19 v2 Artificial Intelligence Machine Learning

Abstract

Through additional training, we explore embedding specialized scientific knowledge into the Llama 2 Large Language Model (LLM). Key findings reveal that effective knowledge integration requires reading texts from multiple perspectives, especially in instructional formats. We utilize text augmentation to tackle the scarcity of specialized texts, including style conversions and translations. Hyperparameter optimization proves crucial, with different size models (7b, 13b, and 70b) reasonably undergoing additional training. Validating our methods, we construct a dataset of 65,000 scientific papers. Although we have succeeded in partially embedding knowledge, the study highlights the complexities and limitations of incorporating specialized information into LLMs, suggesting areas for further improvement.

Keywords

Cite

@article{arxiv.2312.03360,
  title  = {Teaching Specific Scientific Knowledge into Large Language Models through Additional Training},
  author = {Kan Hatakeyama-Sato and Yasuhiko Igarashi and Shun Katakami and Yuta Nabae and Teruaki Hayakawa},
  journal= {arXiv preprint arXiv:2312.03360},
  year   = {2023}
}

Comments

added token information for some texts, and fixed typo

R2 v1 2026-06-28T13:42:36.439Z