Controlling the format of outputs generated by large language models (LLMs) is a critical functionality in various applications. Current methods typically employ constrained decoding with rule-based automata or fine-tuning with manually crafted format instructions, both of which struggle with open-domain format requirements. To address this limitation, we introduce a novel framework for controlled generation in LLMs, leveraging user-provided, one-shot QA pairs. This study investigates LLMs' capabilities to follow open-domain, one-shot constraints and replicate the format of the example answers. We observe that this is a non-trivial problem for current LLMs. We also develop a dataset collection methodology for supervised fine-tuning that enhances the open-domain format control of LLMs without degrading output quality, as well as a benchmark on which we evaluate both the helpfulness and format correctness of LLM outputs. The resulting datasets, named OIFC-SFT, along with the related code, will be made publicly available at https://github.com/cofe-ai/OIFC.
@article{arxiv.2408.04392,
title = {Open-domain Implicit Format Control for Large Language Model Generation},
author = {Yiqun Yao and Wenjia Ma and Xuezhi Fang and Xin Jiang and Xiang Li and Xuying Meng and Peng Han and Jing Li and Aixin Sun and Yequan Wang},
journal= {arXiv preprint arXiv:2408.04392},
year = {2024}
}