@inproceedings{garces-arias-etal-2024-adaptive,
title = "Adaptive Contrastive Search: Uncertainty-Guided Decoding for Open-Ended Text Generation",
author = "Garces Arias, Esteban and
Rodemann, Julian and
Li, Meimingwei and
Heumann, Christian and
A{\ss}enmacher, Matthias",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2024",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-emnlp.885/",
doi = "10.18653/v1/2024.findings-emnlp.885",
pages = "15060--15080",
abstract = "Despite the remarkable capabilities of large language models, generating high-quality text remains a challenging task. Numerous decoding strategies{---}such as beam search, sampling with temperature, top{-}$k$ sampling, nucleus (top{-}$p$) sampling, typical decoding, contrastive decoding, and contrastive search{---}have been proposed to address these challenges by improving coherence, diversity, and resemblance to human-generated text. In this study, we introduce Adaptive Contrastive Search (ACS), a novel decoding strategy that extends contrastive search (CS) by incorporating an adaptive degeneration penalty informed by the model`s estimated uncertainty at each generation step. ACS aims to enhance creativity and diversity while maintaining coherence to produce high-quality outputs. Extensive experiments across various model architectures, languages, and datasets demonstrate that our approach improves both creativity and coherence, underscoring its effectiveness in text-generation tasks. We release our code, datasets, and models to facilitate further research."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="garces-arias-etal-2024-adaptive">
<titleInfo>
<title>Adaptive Contrastive Search: Uncertainty-Guided Decoding for Open-Ended Text Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Esteban</namePart>
<namePart type="family">Garces Arias</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Julian</namePart>
<namePart type="family">Rodemann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Meimingwei</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christian</namePart>
<namePart type="family">Heumann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matthias</namePart>
<namePart type="family">Aßenmacher</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yaser</namePart>
<namePart type="family">Al-Onaizan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohit</namePart>
<namePart type="family">Bansal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yun-Nung</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Miami, Florida, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Despite the remarkable capabilities of large language models, generating high-quality text remains a challenging task. Numerous decoding strategies—such as beam search, sampling with temperature, top-k sampling, nucleus (top-p) sampling, typical decoding, contrastive decoding, and contrastive search—have been proposed to address these challenges by improving coherence, diversity, and resemblance to human-generated text. In this study, we introduce Adaptive Contrastive Search (ACS), a novel decoding strategy that extends contrastive search (CS) by incorporating an adaptive degeneration penalty informed by the model‘s estimated uncertainty at each generation step. ACS aims to enhance creativity and diversity while maintaining coherence to produce high-quality outputs. Extensive experiments across various model architectures, languages, and datasets demonstrate that our approach improves both creativity and coherence, underscoring its effectiveness in text-generation tasks. We release our code, datasets, and models to facilitate further research.</abstract>
<identifier type="citekey">garces-arias-etal-2024-adaptive</identifier>
<identifier type="doi">10.18653/v1/2024.findings-emnlp.885</identifier>
<location>
<url>https://aclanthology.org/2024.findings-emnlp.885/</url>
</location>
<part>
<date>2024-11</date>
<extent unit="page">
<start>15060</start>
<end>15080</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Adaptive Contrastive Search: Uncertainty-Guided Decoding for Open-Ended Text Generation
%A Garces Arias, Esteban
%A Rodemann, Julian
%A Li, Meimingwei
%A Heumann, Christian
%A Aßenmacher, Matthias
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Findings of the Association for Computational Linguistics: EMNLP 2024
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F garces-arias-etal-2024-adaptive
%X Despite the remarkable capabilities of large language models, generating high-quality text remains a challenging task. Numerous decoding strategies—such as beam search, sampling with temperature, top-k sampling, nucleus (top-p) sampling, typical decoding, contrastive decoding, and contrastive search—have been proposed to address these challenges by improving coherence, diversity, and resemblance to human-generated text. In this study, we introduce Adaptive Contrastive Search (ACS), a novel decoding strategy that extends contrastive search (CS) by incorporating an adaptive degeneration penalty informed by the model‘s estimated uncertainty at each generation step. ACS aims to enhance creativity and diversity while maintaining coherence to produce high-quality outputs. Extensive experiments across various model architectures, languages, and datasets demonstrate that our approach improves both creativity and coherence, underscoring its effectiveness in text-generation tasks. We release our code, datasets, and models to facilitate further research.
%R 10.18653/v1/2024.findings-emnlp.885
%U https://aclanthology.org/2024.findings-emnlp.885/
%U https://doi.org/10.18653/v1/2024.findings-emnlp.885
%P 15060-15080
Markdown (Informal)
[Adaptive Contrastive Search: Uncertainty-Guided Decoding for Open-Ended Text Generation](https://aclanthology.org/2024.findings-emnlp.885/) (Garces Arias et al., Findings 2024)
ACL