MetaMetrics-MT: Tuning Meta-Metrics for Machine Translation via Human Preference Calibration

We present MetaMetrics-MT, an innovative metric designed to evaluate machine translation (MT) tasks by aligning closely with human preferences through Bayesian optimization with Gaussian Processes. MetaMetrics-MT enhances existing MT metrics by optimizing their correlation with human judgments. Our...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Anugraha, David, Kuwanto, Garry, Susanto, Lucky, Wijaya, Derry Tanti, Winata, Genta Indra
Format:	Artikel
Sprache:	eng
Schlagworte:	Computer Science - Artificial Intelligence Computer Science - Computation and Language Computer Science - Learning
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

container_end_page
container_issue
container_start_page
container_title
container_volume
creator	Anugraha, David Kuwanto, Garry Susanto, Lucky Wijaya, Derry Tanti Winata, Genta Indra
description	We present MetaMetrics-MT, an innovative metric designed to evaluate machine translation (MT) tasks by aligning closely with human preferences through Bayesian optimization with Gaussian Processes. MetaMetrics-MT enhances existing MT metrics by optimizing their correlation with human judgments. Our experiments on the WMT24 metric shared task dataset demonstrate that MetaMetrics-MT outperforms all existing baselines, setting a new benchmark for state-of-the-art performance in the reference-based setting. Furthermore, it achieves comparable results to leading metrics in the reference-free setting, offering greater efficiency.
doi_str_mv	10.48550/arxiv.2411.00390
format	Article
fullrecord	<record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2411_00390</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2411_00390</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2411_003903</originalsourceid><addsrcrecordid>eNqFjs0KglAQRu-mRVQP0Kp5Ae2aCtVWCjdCi7sMZJKxBnSM8Yd6-1Lat_j44HAWx5h1YP1oH8d2i_riwd9FQeBbGx7s3Fwz6vA75aL1MncE1wvLHUbs_TiUjUKGxYOFwClKW2HHjcDACGlfo8BFqSQlKQgSrPimk7A0sxKrlla_X5jN-eSS1Js68qdyjfrOx5586gn_Gx_sDkB9</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>MetaMetrics-MT: Tuning Meta-Metrics for Machine Translation via Human Preference Calibration</title><source>arXiv.org</source><creator>Anugraha, David ; Kuwanto, Garry ; Susanto, Lucky ; Wijaya, Derry Tanti ; Winata, Genta Indra</creator><creatorcontrib>Anugraha, David ; Kuwanto, Garry ; Susanto, Lucky ; Wijaya, Derry Tanti ; Winata, Genta Indra</creatorcontrib><description>We present MetaMetrics-MT, an innovative metric designed to evaluate machine translation (MT) tasks by aligning closely with human preferences through Bayesian optimization with Gaussian Processes. MetaMetrics-MT enhances existing MT metrics by optimizing their correlation with human judgments. Our experiments on the WMT24 metric shared task dataset demonstrate that MetaMetrics-MT outperforms all existing baselines, setting a new benchmark for state-of-the-art performance in the reference-based setting. Furthermore, it achieves comparable results to leading metrics in the reference-free setting, offering greater efficiency.</description><identifier>DOI: 10.48550/arxiv.2411.00390</identifier><language>eng</language><subject>Computer Science - Artificial Intelligence ; Computer Science - Computation and Language ; Computer Science - Learning</subject><creationdate>2024-11</creationdate><rights>http://creativecommons.org/licenses/by/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2411.00390$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2411.00390$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Anugraha, David</creatorcontrib><creatorcontrib>Kuwanto, Garry</creatorcontrib><creatorcontrib>Susanto, Lucky</creatorcontrib><creatorcontrib>Wijaya, Derry Tanti</creatorcontrib><creatorcontrib>Winata, Genta Indra</creatorcontrib><title>MetaMetrics-MT: Tuning Meta-Metrics for Machine Translation via Human Preference Calibration</title><description>We present MetaMetrics-MT, an innovative metric designed to evaluate machine translation (MT) tasks by aligning closely with human preferences through Bayesian optimization with Gaussian Processes. MetaMetrics-MT enhances existing MT metrics by optimizing their correlation with human judgments. Our experiments on the WMT24 metric shared task dataset demonstrate that MetaMetrics-MT outperforms all existing baselines, setting a new benchmark for state-of-the-art performance in the reference-based setting. Furthermore, it achieves comparable results to leading metrics in the reference-free setting, offering greater efficiency.</description><subject>Computer Science - Artificial Intelligence</subject><subject>Computer Science - Computation and Language</subject><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNqFjs0KglAQRu-mRVQP0Kp5Ae2aCtVWCjdCi7sMZJKxBnSM8Yd6-1Lat_j44HAWx5h1YP1oH8d2i_riwd9FQeBbGx7s3Fwz6vA75aL1MncE1wvLHUbs_TiUjUKGxYOFwClKW2HHjcDACGlfo8BFqSQlKQgSrPimk7A0sxKrlla_X5jN-eSS1Js68qdyjfrOx5586gn_Gx_sDkB9</recordid><startdate>20241101</startdate><enddate>20241101</enddate><creator>Anugraha, David</creator><creator>Kuwanto, Garry</creator><creator>Susanto, Lucky</creator><creator>Wijaya, Derry Tanti</creator><creator>Winata, Genta Indra</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20241101</creationdate><title>MetaMetrics-MT: Tuning Meta-Metrics for Machine Translation via Human Preference Calibration</title><author>Anugraha, David ; Kuwanto, Garry ; Susanto, Lucky ; Wijaya, Derry Tanti ; Winata, Genta Indra</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2411_003903</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Artificial Intelligence</topic><topic>Computer Science - Computation and Language</topic><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Anugraha, David</creatorcontrib><creatorcontrib>Kuwanto, Garry</creatorcontrib><creatorcontrib>Susanto, Lucky</creatorcontrib><creatorcontrib>Wijaya, Derry Tanti</creatorcontrib><creatorcontrib>Winata, Genta Indra</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Anugraha, David</au><au>Kuwanto, Garry</au><au>Susanto, Lucky</au><au>Wijaya, Derry Tanti</au><au>Winata, Genta Indra</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>MetaMetrics-MT: Tuning Meta-Metrics for Machine Translation via Human Preference Calibration</atitle><date>2024-11-01</date><risdate>2024</risdate><abstract>We present MetaMetrics-MT, an innovative metric designed to evaluate machine translation (MT) tasks by aligning closely with human preferences through Bayesian optimization with Gaussian Processes. MetaMetrics-MT enhances existing MT metrics by optimizing their correlation with human judgments. Our experiments on the WMT24 metric shared task dataset demonstrate that MetaMetrics-MT outperforms all existing baselines, setting a new benchmark for state-of-the-art performance in the reference-based setting. Furthermore, it achieves comparable results to leading metrics in the reference-free setting, offering greater efficiency.</abstract><doi>10.48550/arxiv.2411.00390</doi><oa>free_for_read</oa></addata></record>
fulltext	fulltext_linktorsrc
identifier	DOI: 10.48550/arxiv.2411.00390
ispartof
issn
language	eng
recordid	cdi_arxiv_primary_2411_00390
source	arXiv.org
subjects	Computer Science - Artificial Intelligence Computer Science - Computation and Language Computer Science - Learning
title	MetaMetrics-MT: Tuning Meta-Metrics for Machine Translation via Human Preference Calibration
url	https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-27T20%3A17%3A42IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=MetaMetrics-MT:%20Tuning%20Meta-Metrics%20for%20Machine%20Translation%20via%20Human%20Preference%20Calibration&rft.au=Anugraha,%20David&rft.date=2024-11-01&rft_id=info:doi/10.48550/arxiv.2411.00390&rft_dat=%3Carxiv_GOX%3E2411_00390%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true