“Task-Aware Regularized Continual Fine-Tuning for Mitigating Catastrophic Forgetting in Large Language Models” (2025) Journal of Computer Technology and Software, 4(7). doi:10.5281/zenodo.21528630.