Co-evolved Self-Critique: Enhancing Large Language Models with Self-Generated Data
Keywords: large language models, self-critique, co-evolution
TL;DR: We propose a co-evolved self-critique framework for LLMs that improves both generation and evaluation processes, offering a scalable solution to ensure high-quality self-generated data.
Abstract: Large language models (LLMs) have seen staggering progress in recent years. Contemporary LLMs rely on an immense amount of data for training, however, as LLMs continue to advance, the availability of high-quality external data is reaching a bottleneck, highlighting the need for model-generated data for further improvement. Although promising, directly utilizing the self-generated data for model training without scrutinized assessment or filtering can easily lead to deteriorated performance, or in other words, "garbage in, garbage out". In this study, our insight is to carefully craft a \textit{self-critique} process, by equipping the LLMs with the ability to be self-aware and discriminative to the quality of its generated data. We introduce a co-evolved self-critique framework that enables an LLM to simultaneously enhance both its generative and evaluative capabilities through an iterative training process. This provides a scalable solution to ensure high-quality self-generated data and facilitate sustained model improvement. Fine-tuning Llama-3 models using this framework results in encouraging improvements in both instruction-following and discriminative abilities, demonstrating the effectiveness of our method.
Primary Area: foundation or frontier models, including LLMs
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.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
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: 9791
Loading