@inproceedings{shulev-simaan-2024-continual,
title = "Continual Reinforcement Learning for Controlled Text Generation",
author = "Shulev, Velizar and
Sima{'}an, Khalil",
editor = "Calzolari, Nicoletta and
Kan, Min-Yen and
Hoste, Veronique and
Lenci, Alessandro and
Sakti, Sakriani and
Xue, Nianwen",
booktitle = "Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)",
month = may,
year = "2024",
address = "Torino, Italia",
publisher = "ELRA and ICCL",
url = "https://aclanthology.org/2024.lrec-main.343/",
pages = "3881--3889",
abstract = "Controlled Text Generation (CTG) steers the generation of continuations of a given context (prompt) by a Large Language Model (LLM) towards texts possessing a given attribute (e.g., topic, sentiment). In this paper we view CTG as a Continual Learning problem: how to learn at every step to steer next-word generation, without having to wait for end-of-sentence. This continual view is useful for online applications such as CTG for speech, where end-of-sentence is often uncertain. We depart from an existing model, the Plug-and-Play language models (PPLM), which perturbs the context at each step to better predict next-words that posses the desired attribute. While PPLM is intricate and has many hyper-parameters, we provide a proof that the PPLM objective function can be reduced to a Continual Reinforcement Learning (CRL) reward function, thereby simplifying PPLM and endowing it with a better understood learning framework. Subsequently, we present, the first of its kind, CTG algorithm that is fully based on CRL and exhibit promising empirical results."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="shulev-simaan-2024-continual">
<titleInfo>
<title>Continual Reinforcement Learning for Controlled Text Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Velizar</namePart>
<namePart type="family">Shulev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Khalil</namePart>
<namePart type="family">Sima’an</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nicoletta</namePart>
<namePart type="family">Calzolari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Min-Yen</namePart>
<namePart type="family">Kan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Veronique</namePart>
<namePart type="family">Hoste</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alessandro</namePart>
<namePart type="family">Lenci</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakriani</namePart>
<namePart type="family">Sakti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nianwen</namePart>
<namePart type="family">Xue</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA and ICCL</publisher>
<place>
<placeTerm type="text">Torino, Italia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Controlled Text Generation (CTG) steers the generation of continuations of a given context (prompt) by a Large Language Model (LLM) towards texts possessing a given attribute (e.g., topic, sentiment). In this paper we view CTG as a Continual Learning problem: how to learn at every step to steer next-word generation, without having to wait for end-of-sentence. This continual view is useful for online applications such as CTG for speech, where end-of-sentence is often uncertain. We depart from an existing model, the Plug-and-Play language models (PPLM), which perturbs the context at each step to better predict next-words that posses the desired attribute. While PPLM is intricate and has many hyper-parameters, we provide a proof that the PPLM objective function can be reduced to a Continual Reinforcement Learning (CRL) reward function, thereby simplifying PPLM and endowing it with a better understood learning framework. Subsequently, we present, the first of its kind, CTG algorithm that is fully based on CRL and exhibit promising empirical results.</abstract>
<identifier type="citekey">shulev-simaan-2024-continual</identifier>
<location>
<url>https://aclanthology.org/2024.lrec-main.343/</url>
</location>
<part>
<date>2024-05</date>
<extent unit="page">
<start>3881</start>
<end>3889</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Continual Reinforcement Learning for Controlled Text Generation
%A Shulev, Velizar
%A Sima’an, Khalil
%Y Calzolari, Nicoletta
%Y Kan, Min-Yen
%Y Hoste, Veronique
%Y Lenci, Alessandro
%Y Sakti, Sakriani
%Y Xue, Nianwen
%S Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
%D 2024
%8 May
%I ELRA and ICCL
%C Torino, Italia
%F shulev-simaan-2024-continual
%X Controlled Text Generation (CTG) steers the generation of continuations of a given context (prompt) by a Large Language Model (LLM) towards texts possessing a given attribute (e.g., topic, sentiment). In this paper we view CTG as a Continual Learning problem: how to learn at every step to steer next-word generation, without having to wait for end-of-sentence. This continual view is useful for online applications such as CTG for speech, where end-of-sentence is often uncertain. We depart from an existing model, the Plug-and-Play language models (PPLM), which perturbs the context at each step to better predict next-words that posses the desired attribute. While PPLM is intricate and has many hyper-parameters, we provide a proof that the PPLM objective function can be reduced to a Continual Reinforcement Learning (CRL) reward function, thereby simplifying PPLM and endowing it with a better understood learning framework. Subsequently, we present, the first of its kind, CTG algorithm that is fully based on CRL and exhibit promising empirical results.
%U https://aclanthology.org/2024.lrec-main.343/
%P 3881-3889
Markdown (Informal)
[Continual Reinforcement Learning for Controlled Text Generation](https://aclanthology.org/2024.lrec-main.343/) (Shulev & Sima’an, LREC-COLING 2024)
ACL