@inproceedings{zhang-etal-2024-eplsa,
title = "{E}p{LSA}: Synergy of Expert-prefix Mixtures and Task-Oriented Latent Space Adaptation for Diverse Generative Reasoning",
author = "Zhang, Fujun and
Su, Xiangdong and
Li, Jiang and
Yan, Rong and
Gao, Guanglai",
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.549/",
pages = "6216--6227",
abstract = "Existing models for diverse generative reasoning still struggle to generate multiple unique and plausible results. Through an in-depth examination, we argue that it is critical to leverage a mixture of experts as prefixes to enhance the diversity of generated results and make task-oriented adaptation in the latent space of the generation models to improve the quality of the responses. At this point, we propose EpLSA, an innovative model based on the synergy of expert-prefix mixtures and task-oriented latent space adaptation for diverse generative reasoning. Specifically, we use expert-prefixes mixtures to encourage the model to create multiple responses with different semantics and design a loss function to address the problem that the semantics is interfered by the expert-prefixes. Meanwhile, we design a task-oriented adaptation block to make the pre-trained encoder within the generation model more effectively adapted to the pre-trained decoder in the latent space, thus further improving the quality of the generated text. Extensive experiments on three different types of generative reasoning tasks demonstrate that EpLSA outperforms existing baseline models in terms of both the quality and diversity of the generated outputs. Our code is publicly available at https://github.com/IMU-MachineLearningSXD/EpLSA."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2024-eplsa">
<titleInfo>
<title>EpLSA: Synergy of Expert-prefix Mixtures and Task-Oriented Latent Space Adaptation for Diverse Generative Reasoning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Fujun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiangdong</namePart>
<namePart type="family">Su</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiang</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rong</namePart>
<namePart type="family">Yan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Guanglai</namePart>
<namePart type="family">Gao</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>Existing models for diverse generative reasoning still struggle to generate multiple unique and plausible results. Through an in-depth examination, we argue that it is critical to leverage a mixture of experts as prefixes to enhance the diversity of generated results and make task-oriented adaptation in the latent space of the generation models to improve the quality of the responses. At this point, we propose EpLSA, an innovative model based on the synergy of expert-prefix mixtures and task-oriented latent space adaptation for diverse generative reasoning. Specifically, we use expert-prefixes mixtures to encourage the model to create multiple responses with different semantics and design a loss function to address the problem that the semantics is interfered by the expert-prefixes. Meanwhile, we design a task-oriented adaptation block to make the pre-trained encoder within the generation model more effectively adapted to the pre-trained decoder in the latent space, thus further improving the quality of the generated text. Extensive experiments on three different types of generative reasoning tasks demonstrate that EpLSA outperforms existing baseline models in terms of both the quality and diversity of the generated outputs. Our code is publicly available at https://github.com/IMU-MachineLearningSXD/EpLSA.</abstract>
<identifier type="citekey">zhang-etal-2024-eplsa</identifier>
<location>
<url>https://aclanthology.org/2024.lrec-main.549/</url>
</location>
<part>
<date>2024-05</date>
<extent unit="page">
<start>6216</start>
<end>6227</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T EpLSA: Synergy of Expert-prefix Mixtures and Task-Oriented Latent Space Adaptation for Diverse Generative Reasoning
%A Zhang, Fujun
%A Su, Xiangdong
%A Li, Jiang
%A Yan, Rong
%A Gao, Guanglai
%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 zhang-etal-2024-eplsa
%X Existing models for diverse generative reasoning still struggle to generate multiple unique and plausible results. Through an in-depth examination, we argue that it is critical to leverage a mixture of experts as prefixes to enhance the diversity of generated results and make task-oriented adaptation in the latent space of the generation models to improve the quality of the responses. At this point, we propose EpLSA, an innovative model based on the synergy of expert-prefix mixtures and task-oriented latent space adaptation for diverse generative reasoning. Specifically, we use expert-prefixes mixtures to encourage the model to create multiple responses with different semantics and design a loss function to address the problem that the semantics is interfered by the expert-prefixes. Meanwhile, we design a task-oriented adaptation block to make the pre-trained encoder within the generation model more effectively adapted to the pre-trained decoder in the latent space, thus further improving the quality of the generated text. Extensive experiments on three different types of generative reasoning tasks demonstrate that EpLSA outperforms existing baseline models in terms of both the quality and diversity of the generated outputs. Our code is publicly available at https://github.com/IMU-MachineLearningSXD/EpLSA.
%U https://aclanthology.org/2024.lrec-main.549/
%P 6216-6227
Markdown (Informal)
[EpLSA: Synergy of Expert-prefix Mixtures and Task-Oriented Latent Space Adaptation for Diverse Generative Reasoning](https://aclanthology.org/2024.lrec-main.549/) (Zhang et al., LREC-COLING 2024)
ACL