Length Controlled Generation for Black-box LLMs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Gu, Yuxuan, Wang, Wenjie, Feng, Xiaocheng, Zhong, Weihong, Zhu, Kun, Huang, Lei, Chua, Tat-Seng, Qin, Bing
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913618710233088
author Gu, Yuxuan
Wang, Wenjie
Feng, Xiaocheng
Zhong, Weihong
Zhu, Kun
Huang, Lei
Chua, Tat-Seng
Qin, Bing
author_facet Gu, Yuxuan
Wang, Wenjie
Feng, Xiaocheng
Zhong, Weihong
Zhu, Kun
Huang, Lei
Chua, Tat-Seng
Qin, Bing
contents Large language models (LLMs) have demonstrated impressive instruction following capabilities, while still struggling to accurately manage the length of the generated text, which is a fundamental requirement in many real-world applications. Existing length control methods involve fine-tuning the parameters of LLMs, which is inefficient and suboptimal for practical use. In this paper, we propose a novel iterative sampling framework for text length control, integrating the Metropolis-Hastings algorithm with an importance sampling acceleration strategy. This framework efficiently and reliably regulates LLMs to generate length-constrained text without modifying the underlying parameters, thereby preserving the original capabilities of LLMs. Experimental results demonstrate that our framework achieves almost 100\% success rates of length control on Llama3.1 for tasks such as length-controlled abstractive summarization and length-constrained instruction following, with minimal additional computational overhead. This also highlights the significant potential of our method for precise length control across a broader range of applications, without compromising the versatility of LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2412_14656
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Length Controlled Generation for Black-box LLMs
Gu, Yuxuan
Wang, Wenjie
Feng, Xiaocheng
Zhong, Weihong
Zhu, Kun
Huang, Lei
Chua, Tat-Seng
Qin, Bing
Computation and Language
Large language models (LLMs) have demonstrated impressive instruction following capabilities, while still struggling to accurately manage the length of the generated text, which is a fundamental requirement in many real-world applications. Existing length control methods involve fine-tuning the parameters of LLMs, which is inefficient and suboptimal for practical use. In this paper, we propose a novel iterative sampling framework for text length control, integrating the Metropolis-Hastings algorithm with an importance sampling acceleration strategy. This framework efficiently and reliably regulates LLMs to generate length-constrained text without modifying the underlying parameters, thereby preserving the original capabilities of LLMs. Experimental results demonstrate that our framework achieves almost 100\% success rates of length control on Llama3.1 for tasks such as length-controlled abstractive summarization and length-constrained instruction following, with minimal additional computational overhead. This also highlights the significant potential of our method for precise length control across a broader range of applications, without compromising the versatility of LLMs.
title Length Controlled Generation for Black-box LLMs
topic Computation and Language
url https://arxiv.org/abs/2412.14656