Lin, Y., Zhou, Z., Gao, J., Guo, X., Zhang, J., Wu, H., … Wei, X. (2026). GeWu: A Culturally-Grounded Chinese Benchmark for Multi-Stage Social Bias Evaluation in Large Language Models. Proceedings of the AAAI Conference on Artificial Intelligence, 40(38), 32033–32041. https://doi.org/10.1609/aaai.v40i38.40474