{"id":"https://openalex.org/W4412887903","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1011","title":"Exploring Layer-wise Representations of English and Chinese Homonymy in Pre-trained Language Models","display_name":"Exploring Layer-wise Representations of English and Chinese Homonymy in Pre-trained Language Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412887903","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1011"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.1011","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1011","pdf_url":"https://aclanthology.org/2025.findings-acl.1011.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.1011.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5014640347","display_name":"Matthew King-Hang","orcid":"https://orcid.org/0000-0001-9194-2498"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Matthew King-Hang Ma","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Xie Chenwei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie Chenwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100350672","display_name":"Wenbo Wang","orcid":"https://orcid.org/0000-0002-4821-6789"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wenbo Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100702485","display_name":"William Yang Wang","orcid":"https://orcid.org/0000-0001-6153-8240"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"William Shiyuan Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"19705","last_page":"19724"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9972000122070312,"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"}},"topics":[{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9972000122070312,"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":0.9896000027656555,"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/T13910","display_name":"Computational and Text Analysis Methods","score":0.9613000154495239,"subfield":{"id":"https://openalex.org/subfields/3300","display_name":"General Social Sciences"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.7729158997535706},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7443114519119263},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.7041485905647278},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6201223134994507},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.59120774269104},{"id":"https://openalex.org/keywords/philosophy","display_name":"Philosophy","score":0.07171547412872314}],"concepts":[{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.7729158997535706},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7443114519119263},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.7041485905647278},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6201223134994507},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.59120774269104},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.07171547412872314},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.1011","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1011","pdf_url":"https://aclanthology.org/2025.findings-acl.1011.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.1011","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1011","pdf_url":"https://aclanthology.org/2025.findings-acl.1011.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.7200000286102295,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412887903.pdf","grobid_xml":"https://content.openalex.org/works/W4412887903.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Homonymy":[0],"can":[1],"easily":[2],"raise":[3],"lexical":[4],"ambiguity":[5],"due":[6],"to":[7,67,132,182],"the":[8,33,44,55,98,163,184],"misunderstanding":[9],"of":[10,15,36,46,48,57,75,90,100,106,166,186],"its":[11,21],"multiple":[12],"senses.Correct":[13],"recognition":[14],"homonym":[16,52,76,121,129,142],"sense":[17],"greatly":[18],"relies":[19],"on":[20,51],"surrounding":[22],"context.This":[23],"ambiguous":[24],"nature":[25],"makes":[26],"homonyms":[27],"an":[28],"appropriate":[29],"testbed":[30],"for":[31],"examining":[32],"contextualization":[34],"capability":[35],"pre-trained":[37],"(PLM)":[38],"and":[39,54,69,83,104,192],"large":[40],"language":[41],"models":[42,125,145],"(LLMs).Considering":[43],"impact":[45],"part":[47],"speech":[49],"(POS)":[50],"disambiguation":[53,107],"prevalence":[56],"English-focused":[58],"studies":[59],"in":[60,78,119,144,146,159],"word":[61],"embedding":[62],"research,":[63],"this":[64],"study":[65,161],"extends":[66],"Chinese":[68],"provides":[70],"a":[71,101,171],"comprehensive":[72],"layer-wise":[73],"analysis":[74],"representations":[77,130,143],"both":[79],"languages,":[80],"spanning":[81],"same":[82],"different":[84],"POS":[85,140],"categories,":[86],"across":[87],"four":[88],"families":[89],"PLMs/LLMs":[91],"(BERT,":[92],"GPT-2,":[93],"Llama":[94],"3,":[95],"Qwen":[96],"2.5).Through":[97],"creation":[99],"synthetic":[102],"dataset":[103,191],"computation":[105],"score":[108],"(D-Score),":[109],"we":[110],"found":[111],"that:":[112],"(1)":[113],"no":[114],"universal":[115],"layer":[116],"depth":[117],"excels":[118],"differentiating":[120],"representations;":[122],"(2)":[123],"bidirectional":[124],"produce":[126],"better":[127],"contextualized":[128],"compared":[131],"much":[133],"larger":[134],"autoregressive":[135],"models;":[136],"(3)":[137],"most":[138],"importantly,":[139],"affects":[141],"ways":[147],"that":[148],"differ":[149],"from":[150],"human":[151],"research":[152,173],"findings.The":[153],"individual":[154],"differences":[155],"between":[156],"LLMs":[157],"uncovered":[158],"our":[160,190],"challenge":[162],"simplistic":[164],"understanding":[165],"their":[167],"inner":[168],"workings.This":[169],"reveals":[170],"compelling":[172],"frontier:":[174],"conducting":[175],"controlled":[176],"experiments":[177],"with":[178],"purposefully":[179],"manipulated":[180],"inputs":[181],"enhance":[183],"interpretability":[185],"LLMs.We":[187],"have":[188],"made":[189],"codes":[193],"available":[194],"publicly":[195],"at":[196],"https://github.com/neurothew/":[197],"exploring-homonym-rep-in-llm.":[198]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}