Evaluating LLMs’ Capability in Satisfying Lexical Constraints
Bingxuan Li, Yiwei Wang, Tao Meng, Kai-Wei Chang, and Nanyun Peng, in Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2024.
🏆 Best Paper Nomination (2%)
Abstract
This paper analyzes the performance of LLMs in Lexical Constrained Generation (LCG) tasks, identifying key limitations and proposing the Divide and Conquer Generation strategy. Our approach significantly enhances LLMs’ success rate in satisfying lexical constraints across various tasks, providing insights into improving text generation applications.
Bib Entry
@inproceedings{li2024evaluating,
title = {Evaluating LLMs' Capability in Satisfying Lexical Constraints},
author = {Li, Bingxuan and Wang, Yiwei and Meng, Tao and Chang, Kai-Wei and Peng, Nanyun},
booktitle = {Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
year = {2024}
}