AlignDiff: Exploiting Model-Intrinsic Information for Better Preference Data Selection

@article{peng2026,
  title = {AlignDiff: Exploiting Model-Intrinsic Information for Better Preference Data Selection},
  author = {Lai, Peng and Zhu, He and Ruan, Zhiwen and Zhang, Dongdong and Chen, Yun and Li, Peng and Wei, Furu and Liu, Yang and Chen, Guanhua},
  journal = {Findings of the Association for Computational Linguistics: EMNLP 2026},
  year = {2026},
  month = {Jan},
  doi = {10.18653/2026.findings-emnlp.85},
  url = {https://doi.org/10.18653/2026.findings-emnlp.85},
}

Lai, Peng et al. (2026). AlignDiff: Exploiting Model-Intrinsic Information for Better Preference Data Selection. Findings of the Association for Computational Linguistics: EMNLP 2026. https://doi.org/10.18653/2026.findings-emnlp.85

Lai, Peng, et al.. "AlignDiff: Exploiting Model-Intrinsic Information for Better Preference Data Selection." Findings of the Association for Computational Linguistics: EMNLP 2026, 2026, https://doi.org/10.18653/2026.findings-emnlp.85.

Lai, Peng, et al.. "AlignDiff: Exploiting Model-Intrinsic Information for Better Preference Data Selection." Findings of the Association for Computational Linguistics: EMNLP 2026 (2026). https://doi.org/10.18653/2026.findings-emnlp.85

Lai, Peng et al., 2026. AlignDiff: Exploiting Model-Intrinsic Information for Better Preference Data Selection. Findings of the Association for Computational Linguistics: EMNLP 2026. Available at: <https://doi.org/10.18653/2026.findings-emnlp.85>.

TY  - JOUR
TI  - AlignDiff: Exploiting Model-Intrinsic Information for Better Preference Data Selection
AU  - Lai, Peng
AU  - Zhu, He
AU  - Ruan, Zhiwen
AU  - Zhang, Dongdong
AU  - Chen, Yun
AU  - Li, Peng
AU  - Wei, Furu
AU  - Liu, Yang
AU  - Chen, Guanhua
PY  - 2026
JO  - Findings of the Association for Computational Linguistics: EMNLP 2026
DO  - 10.18653/2026.findings-emnlp.85
UR  - https://doi.org/10.18653/2026.findings-emnlp.85
ER  -