@inproceedings{abdollah-pour-etal-2024-gaussian,
title = "{G}aussian Process Optimization for Adaptable Multi-Objective Text Generation using Linearly-Weighted Language Models",
author = "Abdollah Pour, Mohammad Mahdi and
Pesaranghader, Ali and
Cohen, Eldan and
Sanner, Scott",
editor = "Duh, Kevin and
Gomez, Helena and
Bethard, Steven",
booktitle = "Findings of the Association for Computational Linguistics: NAACL 2024",
month = jun,
year = "2024",
address = "Mexico City, Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-naacl.99",
pages = "1529--1536",
abstract = "In multi-objective text generation, we aim to optimize over multiple weighted aspects (e.g., toxicity, semantic preservation, fluency) of the generated text. However, multi-objective weighting schemes may change dynamically in practice according to deployment requirements, evolving business needs, personalization requirements on edge devices, or the availability of new language models and/or objective requirements. Ideally, we need an efficient method to adapt to the dynamic requirements of the overall objective. To address these requirements, we propose a linear combination of objective-specific language models to \textbf{efficiently} adapt the decoding process and optimize for the desired objective \textbf{without} the significant computational overhead of retraining one or more language models. We show empirically that we can leverage Gaussian Process black box optimization to adapt the language model decoder weights to outperform other fixed weighting schemes and standard baselines of the task in only a few iterations of decoding. Overall this approach enables highly efficient adaptation of controllable language models via multi-objective weighting schemes that may evolve dynamically in practical deployment situations.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="abdollah-pour-etal-2024-gaussian">
<titleInfo>
<title>Gaussian Process Optimization for Adaptable Multi-Objective Text Generation using Linearly-Weighted Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Mahdi</namePart>
<namePart type="family">Abdollah Pour</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ali</namePart>
<namePart type="family">Pesaranghader</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eldan</namePart>
<namePart type="family">Cohen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Scott</namePart>
<namePart type="family">Sanner</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-06</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: NAACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kevin</namePart>
<namePart type="family">Duh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Helena</namePart>
<namePart type="family">Gomez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Steven</namePart>
<namePart type="family">Bethard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Mexico City, Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In multi-objective text generation, we aim to optimize over multiple weighted aspects (e.g., toxicity, semantic preservation, fluency) of the generated text. However, multi-objective weighting schemes may change dynamically in practice according to deployment requirements, evolving business needs, personalization requirements on edge devices, or the availability of new language models and/or objective requirements. Ideally, we need an efficient method to adapt to the dynamic requirements of the overall objective. To address these requirements, we propose a linear combination of objective-specific language models to efficiently adapt the decoding process and optimize for the desired objective without the significant computational overhead of retraining one or more language models. We show empirically that we can leverage Gaussian Process black box optimization to adapt the language model decoder weights to outperform other fixed weighting schemes and standard baselines of the task in only a few iterations of decoding. Overall this approach enables highly efficient adaptation of controllable language models via multi-objective weighting schemes that may evolve dynamically in practical deployment situations.</abstract>
<identifier type="citekey">abdollah-pour-etal-2024-gaussian</identifier>
<location>
<url>https://aclanthology.org/2024.findings-naacl.99</url>
</location>
<part>
<date>2024-06</date>
<extent unit="page">
<start>1529</start>
<end>1536</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Gaussian Process Optimization for Adaptable Multi-Objective Text Generation using Linearly-Weighted Language Models
%A Abdollah Pour, Mohammad Mahdi
%A Pesaranghader, Ali
%A Cohen, Eldan
%A Sanner, Scott
%Y Duh, Kevin
%Y Gomez, Helena
%Y Bethard, Steven
%S Findings of the Association for Computational Linguistics: NAACL 2024
%D 2024
%8 June
%I Association for Computational Linguistics
%C Mexico City, Mexico
%F abdollah-pour-etal-2024-gaussian
%X In multi-objective text generation, we aim to optimize over multiple weighted aspects (e.g., toxicity, semantic preservation, fluency) of the generated text. However, multi-objective weighting schemes may change dynamically in practice according to deployment requirements, evolving business needs, personalization requirements on edge devices, or the availability of new language models and/or objective requirements. Ideally, we need an efficient method to adapt to the dynamic requirements of the overall objective. To address these requirements, we propose a linear combination of objective-specific language models to efficiently adapt the decoding process and optimize for the desired objective without the significant computational overhead of retraining one or more language models. We show empirically that we can leverage Gaussian Process black box optimization to adapt the language model decoder weights to outperform other fixed weighting schemes and standard baselines of the task in only a few iterations of decoding. Overall this approach enables highly efficient adaptation of controllable language models via multi-objective weighting schemes that may evolve dynamically in practical deployment situations.
%U https://aclanthology.org/2024.findings-naacl.99
%P 1529-1536
Markdown (Informal)
[Gaussian Process Optimization for Adaptable Multi-Objective Text Generation using Linearly-Weighted Language Models](https://aclanthology.org/2024.findings-naacl.99) (Abdollah Pour et al., Findings 2024)
ACL