{"id":"https://openalex.org/W2512434509","doi":"https://doi.org/10.18653/v1/p16-1182","title":"MUTT: Metric Unit TesTing for Language Generation Tasks","display_name":"MUTT: Metric Unit TesTing for Language Generation Tasks","publication_year":2016,"publication_date":"2016-01-01","ids":{"openalex":"https://openalex.org/W2512434509","doi":"https://doi.org/10.18653/v1/p16-1182","mag":"2512434509"},"language":"en","primary_location":{"id":"doi:10.18653/v1/p16-1182","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/p16-1182","pdf_url":"https://www.aclweb.org/anthology/P16-1182.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.aclweb.org/anthology/P16-1182.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5006156483","display_name":"William Boag","orcid":"https://orcid.org/0000-0002-1485-5806"},"institutions":[{"id":"https://openalex.org/I133738476","display_name":"University of Massachusetts Lowell","ror":"https://ror.org/03hamhx47","country_code":"US","type":"education","lineage":["https://openalex.org/I133738476"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"William Boag","raw_affiliation_strings":["Dept. of Computer Science University of Massachusetts Lowell 198 Riverside St, Lowell, MA 01854"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dept. of Computer Science University of Massachusetts Lowell 198 Riverside St, Lowell, MA 01854","institution_ids":["https://openalex.org/I133738476"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043418731","display_name":"Renan Costa Campos","orcid":"https://orcid.org/0000-0002-3166-2379"},"institutions":[{"id":"https://openalex.org/I133738476","display_name":"University of Massachusetts Lowell","ror":"https://ror.org/03hamhx47","country_code":"US","type":"education","lineage":["https://openalex.org/I133738476"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Renan Campos","raw_affiliation_strings":["Dept. of Computer Science University of Massachusetts Lowell 198 Riverside St, Lowell, MA 01854"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dept. of Computer Science University of Massachusetts Lowell 198 Riverside St, Lowell, MA 01854","institution_ids":["https://openalex.org/I133738476"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075906727","display_name":"Kate Saenko","orcid":"https://orcid.org/0000-0002-7564-7218"},"institutions":[{"id":"https://openalex.org/I133738476","display_name":"University of Massachusetts Lowell","ror":"https://ror.org/03hamhx47","country_code":"US","type":"education","lineage":["https://openalex.org/I133738476"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kate Saenko","raw_affiliation_strings":["Dept. of Computer Science University of Massachusetts Lowell 198 Riverside St, Lowell, MA 01854"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dept. of Computer Science University of Massachusetts Lowell 198 Riverside St, Lowell, MA 01854","institution_ids":["https://openalex.org/I133738476"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5071360545","display_name":"Anna Rumshisky","orcid":null},"institutions":[{"id":"https://openalex.org/I133738476","display_name":"University of Massachusetts Lowell","ror":"https://ror.org/03hamhx47","country_code":"US","type":"education","lineage":["https://openalex.org/I133738476"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Anna Rumshisky","raw_affiliation_strings":["Dept. of Computer Science University of Massachusetts Lowell 198 Riverside St, Lowell, MA 01854"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dept. of Computer Science University of Massachusetts Lowell 198 Riverside St, Lowell, MA 01854","institution_ids":["https://openalex.org/I133738476"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I133738476"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1935","last_page":"1943"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9997000098228455,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9994999766349792,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.8636903762817383},{"id":"https://openalex.org/keywords/sentence","display_name":"Sentence","score":0.7514628171920776},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.7066504955291748},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6998579502105713},{"id":"https://openalex.org/keywords/strengths-and-weaknesses","display_name":"Strengths and weaknesses","score":0.6141703128814697},{"id":"https://openalex.org/keywords/correlation","display_name":"Correlation","score":0.5716129541397095},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5585489869117737},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5585360527038574},{"id":"https://openalex.org/keywords/closed-captioning","display_name":"Closed captioning","score":0.4982888698577881},{"id":"https://openalex.org/keywords/translation","display_name":"Translation (biology)","score":0.4525723159313202},{"id":"https://openalex.org/keywords/simple","display_name":"Simple (philosophy)","score":0.4351913630962372},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.42246949672698975},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.35139888525009155},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.22335901856422424},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.15124091506004333},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.09887704253196716},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.07274466753005981}],"concepts":[{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.8636903762817383},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.7514628171920776},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.7066504955291748},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6998579502105713},{"id":"https://openalex.org/C63882131","wikidata":"https://www.wikidata.org/wiki/Q17122954","display_name":"Strengths and weaknesses","level":2,"score":0.6141703128814697},{"id":"https://openalex.org/C117220453","wikidata":"https://www.wikidata.org/wiki/Q5172842","display_name":"Correlation","level":2,"score":0.5716129541397095},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5585489869117737},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5585360527038574},{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.4982888698577881},{"id":"https://openalex.org/C149364088","wikidata":"https://www.wikidata.org/wiki/Q185917","display_name":"Translation (biology)","level":4,"score":0.4525723159313202},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.4351913630962372},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.42246949672698975},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.35139888525009155},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.22335901856422424},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.15124091506004333},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.09887704253196716},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.07274466753005981},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.0},{"id":"https://openalex.org/C105580179","wikidata":"https://www.wikidata.org/wiki/Q188928","display_name":"Messenger RNA","level":3,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/p16-1182","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/p16-1182","pdf_url":"https://www.aclweb.org/anthology/P16-1182.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/p16-1182","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/p16-1182","pdf_url":"https://www.aclweb.org/anthology/P16-1182.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions","score":0.8199999928474426}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2512434509.pdf","grobid_xml":"https://content.openalex.org/works/W2512434509.grobid-xml"},"referenced_works_count":14,"referenced_works":["https://openalex.org/W1493309689","https://openalex.org/W1889081078","https://openalex.org/W1956340063","https://openalex.org/W2032494091","https://openalex.org/W2101105183","https://openalex.org/W2103305545","https://openalex.org/W2133459682","https://openalex.org/W2149327368","https://openalex.org/W2153579005","https://openalex.org/W2154652894","https://openalex.org/W2159107349","https://openalex.org/W2250790822","https://openalex.org/W2251610689","https://openalex.org/W4294170691"],"related_works":["https://openalex.org/W4210416330","https://openalex.org/W2775506363","https://openalex.org/W3088136942","https://openalex.org/W4290852288","https://openalex.org/W2949362007","https://openalex.org/W4283207562","https://openalex.org/W2963177403","https://openalex.org/W2330246314","https://openalex.org/W2949522393","https://openalex.org/W4388002133"],"abstract_inverted_index":{"Precise":[0],"evaluation":[1,61],"metrics":[2,22,127],"are":[3,34,39,104],"important":[4],"for":[5,59,88],"assessing":[6],"progress":[7],"in":[8,42,109],"high-level":[9],"language":[10],"generation":[11],"tasks":[12],"such":[13],"as":[14,85],"machine":[15],"translation":[16],"or":[17],"image":[18],"captioning.":[19],"Historically,":[20],"these":[21],"have":[23,75],"been":[24],"evaluated":[25],"using":[26],"correlation":[27,138],"with":[28,139],"human":[29,105,140],"judgment.":[30],"However,":[31],"human-derived":[32],"scores":[33],"often":[35],"alarmingly":[36],"inconsistent":[37,108],"and":[38,93,131],"also":[40],"limited":[41],"their":[43],"ability":[44],"to":[45,72,121],"identify":[46],"precise":[47,123],"areas":[48],"of":[49,95,125],"weakness.":[50],"In":[51],"this":[52,110],"paper,":[53],"we":[54],"perform":[55],"a":[56,96,136],"case":[57],"study":[58],"metric":[60,79],"by":[62],"measuring":[63,90],"the":[64,77,91,114],"effect":[65],"that":[66,101,113],"systematic":[67],"sentence":[68,82],"transformations":[69],"(e.g.":[70,128],"active":[71,132],"passive":[73,130],"voice)":[74],"on":[76],"automatic":[78],"scores.":[80],"These":[81],"\"corruptions\"":[83],"serve":[84],"unit":[86],"tests":[87],"precisely":[89],"strengths":[92],"weaknesses":[94],"given":[97],"metric.":[98],"We":[99],"find":[100],"not":[102],"only":[103],"annotations":[106],"heavily":[107],"study,":[111],"but":[112],"Metric":[115],"Unit":[116],"TesT":[117],"analysis":[118],"is":[119],"able":[120],"capture":[122],"shortcomings":[124],"particular":[126],"comparing":[129],"sentences)":[133],"better":[134],"than":[135],"simple":[137],"judgment":[141],"can.":[142]},"counts_by_year":[{"year":2021,"cited_by_count":1},{"year":2020,"cited_by_count":1},{"year":2019,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}