SteerLM: Attribute Conditioned SFT as an (User-Steerable) Alternative to RLHF

Yi Dong, Zhilin Wang, Makesh Narsimhan Sreedhar, Xianchao Wu, Oleksii Kuchaiev. SteerLM: Attribute Conditioned SFT as an (User-Steerable) Alternative to RLHF. In Houda Bouamor, Juan Pino 0001, Kalika Bali, editors, Findings of the Association for Computational Linguistics: EMNLP 2023, Singapore, December 6-10, 2023. pages 11275-11288, Association for Computational Linguistics, 2023. [doi]

@inproceedings{DongWSWK23,
  title = {SteerLM: Attribute Conditioned SFT as an (User-Steerable) Alternative to RLHF},
  author = {Yi Dong and Zhilin Wang and Makesh Narsimhan Sreedhar and Xianchao Wu and Oleksii Kuchaiev},
  year = {2023},
  url = {https://aclanthology.org/2023.findings-emnlp.754},
  researchr = {https://researchr.org/publication/DongWSWK23},
  cites = {0},
  citedby = {0},
  pages = {11275-11288},
  booktitle = {Findings of the Association for Computational Linguistics: EMNLP 2023, Singapore, December 6-10, 2023},
  editor = {Houda Bouamor and Juan Pino 0001 and Kalika Bali},
  publisher = {Association for Computational Linguistics},
  isbn = {979-8-89176-061-5},
}