Control Large Language Models via Divide and Conquer

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Bingxuan, Wang, Yiwei, Meng, Tao, Chang, Kai-Wei, Peng, Nanyun
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916425203974144
author Li, Bingxuan
Wang, Yiwei
Meng, Tao
Chang, Kai-Wei
Peng, Nanyun
author_facet Li, Bingxuan
Wang, Yiwei
Meng, Tao
Chang, Kai-Wei
Peng, Nanyun
contents This paper investigates controllable generation for large language models (LLMs) with prompt-based control, focusing on Lexically Constrained Generation (LCG). We systematically evaluate the performance of LLMs on satisfying lexical constraints with prompt-based control, as well as their efficacy in downstream applications. We conclude that LLMs face significant challenges in consistently satisfying lexical constraints with prompt-based control. We identified three key limitations of LLMs for LCG, including (1) position bias, where LLMs tend to satisfy constraints that appear in specific positions within the input; (2) low responsiveness to decoding parameters, which render minimal impact on control of LLMs; and (3) struggle with handling the inherent complexity of certain constraints (e.g., compound words). To address these issues, we introduce a 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 provides valuable insights into the performance of LLMs in LCG with prompt-based control, and our proposed strategy offers a pathway to more sophisticated and customized text generation applications.
format Preprint
id arxiv_https___arxiv_org_abs_2410_04628
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Control Large Language Models via Divide and Conquer
Li, Bingxuan
Wang, Yiwei
Meng, Tao
Chang, Kai-Wei
Peng, Nanyun
Computation and Language
This paper investigates controllable generation for large language models (LLMs) with prompt-based control, focusing on Lexically Constrained Generation (LCG). We systematically evaluate the performance of LLMs on satisfying lexical constraints with prompt-based control, as well as their efficacy in downstream applications. We conclude that LLMs face significant challenges in consistently satisfying lexical constraints with prompt-based control. We identified three key limitations of LLMs for LCG, including (1) position bias, where LLMs tend to satisfy constraints that appear in specific positions within the input; (2) low responsiveness to decoding parameters, which render minimal impact on control of LLMs; and (3) struggle with handling the inherent complexity of certain constraints (e.g., compound words). To address these issues, we introduce a 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 provides valuable insights into the performance of LLMs in LCG with prompt-based control, and our proposed strategy offers a pathway to more sophisticated and customized text generation applications.
title Control Large Language Models via Divide and Conquer
topic Computation and Language
url https://arxiv.org/abs/2410.04628