M. Thelwall, “Can smaller large language models evaluate research quality?”, MJLIS, vol. 30, no. 2, pp. 66–81, Aug. 2025.