ReQuEST: A Small-Scale Multi-Task Model for Community Question-Answering Systems
The burgeoning popularity of community question-answering platforms as an information-seeking strategy has prompted researchers to look for ways to save response time and effort, among which question entailment recognizing, question summarizing, and question tagging are prominent. However, none has...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2024-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/10413543/ |
_version_ | 1797323892666138624 |
---|---|
author | Seyyede Zahra Aftabi Seyyede Maryam Seyyedi Mohammad Maleki Saeed Farzi |
author_facet | Seyyede Zahra Aftabi Seyyede Maryam Seyyedi Mohammad Maleki Saeed Farzi |
author_sort | Seyyede Zahra Aftabi |
collection | DOAJ |
description | The burgeoning popularity of community question-answering platforms as an information-seeking strategy has prompted researchers to look for ways to save response time and effort, among which question entailment recognizing, question summarizing, and question tagging are prominent. However, none has investigated the implicit relations between these tasks and the benefits their interaction could provide. In this study, ReQuEST, a novel multi-task model based on bidirectional auto-regressive transformers (BART), is introduced to recognize question entailment, summarize questions respecting given queries, and tag questions with primary topics, simultaneously. ReQuEST comprises one shared encoder representing input sequences, two half-shared decoders providing intermediate presentations, and three task-specific heads producing summaries, tags, and entailed questions. A lightweight fine-tuning technique and a weighted loss function help us learn model parameters efficiently. With roughly 187k learning parameters, ReQuEST is almost half the size of BARTlarge and is two-thirds smaller than its multi-task counterparts. Empirical experiments on standard summarization datasets reveal that ReQuEST outperforms competitors on Debatepedia with a Rouge-L of 46.77 and has persuasive performance with a Rouge-L of 37.37 on MeQSum. On MediQA-RQE as a medical benchmark for entailment recognition, ReQuEST is also comparable in accuracy with state-of-the-art systems without being pre-trained on domain-specific datasets. |
first_indexed | 2024-03-08T05:34:41Z |
format | Article |
id | doaj.art-2139a6b41c164c68b0a01bf25509864f |
institution | Directory Open Access Journal |
issn | 2169-3536 |
language | English |
last_indexed | 2024-03-08T05:34:41Z |
publishDate | 2024-01-01 |
publisher | IEEE |
record_format | Article |
series | IEEE Access |
spelling | doaj.art-2139a6b41c164c68b0a01bf25509864f2024-02-06T00:01:03ZengIEEEIEEE Access2169-35362024-01-0112171371715110.1109/ACCESS.2024.335828710413543ReQuEST: A Small-Scale Multi-Task Model for Community Question-Answering SystemsSeyyede Zahra Aftabi0https://orcid.org/0000-0002-3651-9723Seyyede Maryam Seyyedi1https://orcid.org/0000-0002-3653-9214Mohammad Maleki2https://orcid.org/0000-0001-5726-6596Saeed Farzi3https://orcid.org/0000-0003-2850-0616Faculty of Computer Engineering, K. N. Toosi University of Technology, Tehran, IranFaculty of Computer Engineering, K. N. Toosi University of Technology, Tehran, IranFaculty of Computer Engineering, K. N. Toosi University of Technology, Tehran, IranFaculty of Computer Engineering, K. N. Toosi University of Technology, Tehran, IranThe burgeoning popularity of community question-answering platforms as an information-seeking strategy has prompted researchers to look for ways to save response time and effort, among which question entailment recognizing, question summarizing, and question tagging are prominent. However, none has investigated the implicit relations between these tasks and the benefits their interaction could provide. In this study, ReQuEST, a novel multi-task model based on bidirectional auto-regressive transformers (BART), is introduced to recognize question entailment, summarize questions respecting given queries, and tag questions with primary topics, simultaneously. ReQuEST comprises one shared encoder representing input sequences, two half-shared decoders providing intermediate presentations, and three task-specific heads producing summaries, tags, and entailed questions. A lightweight fine-tuning technique and a weighted loss function help us learn model parameters efficiently. With roughly 187k learning parameters, ReQuEST is almost half the size of BARTlarge and is two-thirds smaller than its multi-task counterparts. Empirical experiments on standard summarization datasets reveal that ReQuEST outperforms competitors on Debatepedia with a Rouge-L of 46.77 and has persuasive performance with a Rouge-L of 37.37 on MeQSum. On MediQA-RQE as a medical benchmark for entailment recognition, ReQuEST is also comparable in accuracy with state-of-the-art systems without being pre-trained on domain-specific datasets.https://ieeexplore.ieee.org/document/10413543/Community question answering systemsmulti-task learningquery-focused question summarizationquestion entailmenttag generation |
spellingShingle | Seyyede Zahra Aftabi Seyyede Maryam Seyyedi Mohammad Maleki Saeed Farzi ReQuEST: A Small-Scale Multi-Task Model for Community Question-Answering Systems IEEE Access Community question answering systems multi-task learning query-focused question summarization question entailment tag generation |
title | ReQuEST: A Small-Scale Multi-Task Model for Community Question-Answering Systems |
title_full | ReQuEST: A Small-Scale Multi-Task Model for Community Question-Answering Systems |
title_fullStr | ReQuEST: A Small-Scale Multi-Task Model for Community Question-Answering Systems |
title_full_unstemmed | ReQuEST: A Small-Scale Multi-Task Model for Community Question-Answering Systems |
title_short | ReQuEST: A Small-Scale Multi-Task Model for Community Question-Answering Systems |
title_sort | request a small scale multi task model for community question answering systems |
topic | Community question answering systems multi-task learning query-focused question summarization question entailment tag generation |
url | https://ieeexplore.ieee.org/document/10413543/ |
work_keys_str_mv | AT seyyedezahraaftabi requestasmallscalemultitaskmodelforcommunityquestionansweringsystems AT seyyedemaryamseyyedi requestasmallscalemultitaskmodelforcommunityquestionansweringsystems AT mohammadmaleki requestasmallscalemultitaskmodelforcommunityquestionansweringsystems AT saeedfarzi requestasmallscalemultitaskmodelforcommunityquestionansweringsystems |