Data-Centric Human Preference Optimization with Rationales
Keywords: dpo, preference learning, alignment
TL;DR: Human Preference Optimization with Rationales
Abstract: Reinforcement learning from human feedback plays a crucial role in aligning
language models towards human preferences, traditionally represented through
comparisons between pairs or sets of responses within a given context. While
many studies have enhanced algorithmic techniques to optimize learning from such
data, this work shifts focus to improving preference learning through a data-centric
approach. Specifically, we propose enriching existing preference datasets with
machine-generated rationales that explain the reasons behind choices. We develop
a simple and principled framework to augment current preference learning methods
with rationale information. Our comprehensive analysis highlights how rationales
enhance learning efficiency. Extensive experiments reveal that rationale-enriched
preference learning offers multiple advantages: it improves annotation efficiency,
accelerates convergence to higher-performing models, and reduces verbosity bias
and hallucination. Furthermore, this framework is versatile enough to integrate
with various preference optimization algorithms. Overall, our findings highlight
the potential of re-imagining data design for preference learning, demonstrating
that even freely available machine-generated rationales can significantly boost
performance across multiple dimensions.
Primary Area: alignment, fairness, safety, privacy, and societal considerations
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 8553
Loading