{"id":"https://openalex.org/W4412945438","doi":"https://doi.org/10.18653/v1/2025.acl-long.573","title":"Less Mature is More Adaptable for Sentence-level Language Modeling","display_name":"Less Mature is More Adaptable for Sentence-level Language Modeling","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412945438","doi":"https://doi.org/10.18653/v1/2025.acl-long.573"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.573","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.573","pdf_url":"https://aclanthology.org/2025.acl-long.573.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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.573.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5030810711","display_name":"Abhilasha Sancheti","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Abhilasha Sancheti","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015520404","display_name":"David C. Dale","orcid":"https://orcid.org/0000-0002-7146-1440"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"David Dale","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112988007","display_name":"Artyom Kozhevnikov","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Artyom Kozhevnikov","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5005764737","display_name":"Maha Elbayad","orcid":"https://orcid.org/0000-0002-8389-231X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Maha Elbayad","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.08887238,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11680","last_page":"11695"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9340000152587891,"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/T10028","display_name":"Topic Modeling","score":0.9340000152587891,"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/computer-science","display_name":"Computer science","score":0.7750760316848755},{"id":"https://openalex.org/keywords/sentence","display_name":"Sentence","score":0.5622281432151794},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.49412277340888977},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4100003242492676},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.34091150760650635}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7750760316848755},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.5622281432151794},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.49412277340888977},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4100003242492676},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.34091150760650635}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.573","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.573","pdf_url":"https://aclanthology.org/2025.acl-long.573.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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.573","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.573","pdf_url":"https://aclanthology.org/2025.acl-long.573.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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.7400000095367432,"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/W4412945438.pdf","grobid_xml":"https://content.openalex.org/works/W4412945438.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":{"This":[0],"work":[1],"investigates":[2],"sentence-level":[3,144],"models":[4,6,104],"(i.e.,":[5],"that":[7,70,122],"operate":[8],"at":[9],"the":[10,138],"sentence-level)":[11],"to":[12,53,95,99],"study":[13],"how":[14],"sentence":[15,58],"representations":[16,73,76,84,100],"from":[17,77,85,101],"various":[18,49],"encoders":[19,38,86],"influence":[20],"downstream":[21,96,131],"task":[22],"performance,":[23],"and":[24,28,43,62,93,124,140],"which":[25],"syntactic,":[26],"semantic,":[27],"discourse-level":[29,125],"properties":[30,126],"are":[31,110,127],"essential":[32],"for":[33,117,153],"strong":[34],"performance.Our":[35],"experiments":[36],"encompass":[37],"with":[39,87],"diverse":[40],"training":[41],"regimes":[42],"pretraining":[44,114],"domains,":[45],"as":[46,48],"well":[47],"pooling":[50],"strategies":[51],"applied":[52],"multi-sentence":[54],"input":[55],"tasks":[56,97],"(including":[57],"ordering,":[59],"sentiment":[60],"classification,":[61],"natural":[63],"language":[64],"inference)":[65],"requiring":[66],"coarse-to-fine-grained":[67],"reasoning.We":[68],"find":[69],"\"less":[71],"mature\"":[72],"(e.g.,":[74],"mean-pooled":[75],"BERT's":[78],"first":[79],"or":[80,83,107,136],"last":[81],"layer,":[82],"limited":[88],"fine-tuning)":[89],"exhibit":[90],"greater":[91],"generalizability":[92],"adaptability":[94],"compared":[98],"extensively":[102],"fine-tuned":[103],"(e.g.,,":[105],"SBERT":[106],"Sim-CSE).These":[108],"findings":[109],"consistent":[111],"across":[112],"different":[113],"seed":[115],"initializations":[116],"BERT.Our":[118],"probing":[119],"analysis":[120],"reveals":[121],"syntactic":[123],"stronger":[128],"indicators":[129],"of":[130,143],"performance":[132],"than":[133],"MTEB":[134],"scores":[135],"decodability.Furthermore,":[137],"data":[139],"time":[141],"efficiency":[142],"models,":[145,149],"often":[146],"outperforming":[147],"token-level":[148],"underscores":[150],"their":[151],"potential":[152],"future":[154],"research.":[155]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}