@article{alignmentweighteddpoaprincipledreasoning, title = {Alignment-Weighted DPO: A principled reasoning approach to improve safety alignment}, author = {Mengxuan Hu and Vivek V. Datla and Anoop Kumar and Zihan Guan and Sheng Li and Alfy Samuel and Daben Liu}, year = {2026}, eprint = {2602.21346}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2602.21346}, }