IOPO: Empowering LLMs with Complex Instruction Following via Input-Output Preference Optimization.

Xinghua Zhang, Haiyang Yu, Cheng Fu, Fei Huang, Yongbin Li

Browse the full ACL paper archive.