Control Large Language Models via Divide and Conquer

Bingxuan Li, Yiwei Wang, Tao Meng, Kai-Wei Chang, Nanyun Peng


Abstract
This paper investigates the capability of LLMs on controllable generation with prompt-based controlling, focusing on Lexically Constrained Generation (LCG). We systematically evaluate the performance of LLMs on satisfying lexical constraints with prompt-based controlling, as well as their efficacy in downstream applications. We identified three key reasons that highlight the limitations of LLMs in LCG, including (1) position bias, where LLMs tend to satisfy constraints that appear in specific positions within the input; (2) low responsiveness to control decoding parameters, which minimally impact the performance of LLMs; and (3) struggle with handling the inherent complexity of certain constraints (e.g. compound word). We conclude that black-box LLMs face significant challenges in consistently satisfying lexical constraints with prompt-based controlling. To address this bottleneck, we introduce the Divide and Conquer Generation strategy, effective for both white-box and black-box LLMs, to enhance LLMs performance in LCG tasks, which demonstrates over 90% improvement on success rate in the most challenging LCG task. Our analysis aims to provide valuable insights into the performance of LLMs in LCG with prompt-based controlling, and our proposed strategy offers a pathway to more sophisticated and customized text generation applications.
Anthology ID:
2024.emnlp-main.850
Volume:
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
15240–15256
Language:
URL:
https://aclanthology.org/2024.emnlp-main.850
DOI:
Bibkey:
Cite (ACL):
Bingxuan Li, Yiwei Wang, Tao Meng, Kai-Wei Chang, and Nanyun Peng. 2024. Control Large Language Models via Divide and Conquer. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 15240–15256, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
Control Large Language Models via Divide and Conquer (Li et al., EMNLP 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.emnlp-main.850.pdf