Extrapolating to Unknown Opinions Using LLMs
Kexun Zhang, Jane Dwivedi-Yu, Zhaojiang Lin, Yuning Mao, William Yang Wang, Lei Li, Yi-Chia Wang
Abstract
From ice cream flavors to climate change, people exhibit a wide array of opinions on various topics, and understanding the rationale for these opinions can promote healthy discussion and consensus among them. As such, it can be valuable for a large language model (LLM), particularly as an AI assistant, to be able to empathize with or even explain these various standpoints. In this work, we hypothesize that different topic stances often manifest correlations that can be used to extrapolate to topics with unknown opinions. We explore various prompting and fine-tuning methods to improve an LLM’s ability to (a) extrapolate from opinions on known topics to unknown ones and (b) support their extrapolation with reasoning. Our findings suggest that LLMs possess inherent knowledge from training data about these opinion correlations, and with minimal data, the similarities between human opinions and model-extrapolated opinions can be improved by more than 50%. Furthermore, LLM can generate the reasoning process behind their extrapolation of opinions.
BibTeX
@inproceedings{zhang-etal-2025-extrapolating,
title = "Extrapolating to Unknown Opinions Using {LLM}s",
author = "Zhang, Kexun and
Dwivedi-Yu, Jane and
Lin, Zhaojiang and
Mao, Yuning and
Wang, William Yang and
Li, Lei and
Wang, Yi-Chia",
editor = "Rambow, Owen and
Wanner, Leo and
Apidianaki, Marianna and
Al-Khalifa, Hend and
Eugenio, Barbara Di and
Schockaert, Steven",
booktitle = "Proceedings of the 31st International Conference on Computational Linguistics",
month = jan,
year = "2025",
address = "Abu Dhabi, UAE",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.coling-main.523/",
pages = "7819--7830"
}