Arwa Ahmed Mohamed Ahmed Badawy

Vertical Scaling of Subdomain-Specific Small Language Models in Complex Aerospace Engineering: Synthetic QA Generation, Fine-Tuning, and LLM-as-a-Judge Evaluation

2026 · Zenodo (CERN European Organization for Nuclear Research)

Large Language Models (LLMs) demonstrate strong general-purpose reasoning, but their high inference cost and latency limit deployment in specialized engineering contexts. This paper investigates whether a localized, subdomain-adapted Small Language Model (SLM) can achieve competitive …