Liu, R. et al. 2026. SafeNLIDB: A Privacy-Preserving Safety Alignment Framework for LLM-based Natural Language Database Interfaces. Proceedings of the AAAI Conference on Artificial Intelligence. 40, 38 (Mar. 2026), 32123–32131. DOI:https://doi.org/10.1609/aaai.v40i38.40484.