{"id":"https://openalex.org/W4388778142","doi":"https://doi.org/10.1162/tacl_a_00608","title":"How Abstract Is Linguistic Generalization in Large Language Models? Experiments with Argument Structure","display_name":"How Abstract Is Linguistic Generalization in Large Language Models? Experiments with Argument Structure","publication_year":2023,"publication_date":"2023-01-01","ids":{"openalex":"https://openalex.org/W4388778142","doi":"https://doi.org/10.1162/tacl_a_00608"},"language":"en","primary_location":{"id":"doi:10.1162/tacl_a_00608","doi":"https://doi.org/10.1162/tacl_a_00608","is_oa":true,"landing_page_url":"https://doi.org/10.1162/tacl_a_00608","pdf_url":"https://direct.mit.edu/tacl/article-pdf/doi/10.1162/tacl_a_00608/2178846/tacl_a_00608.pdf","source":{"id":"https://openalex.org/S2729999759","display_name":"Transactions of the Association for Computational Linguistics","issn_l":"2307-387X","issn":["2307-387X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["ccf-b","cwts-core","doaj","dongbi-b","jufo-1","norway-2","poland-20","russia-white-list-1"],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Transactions of the Association for Computational Linguistics","raw_type":"journal-article"},"volume":null,"type":"article","indexed_in":["arxiv","crossref","datacite","doaj"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://direct.mit.edu/tacl/article-pdf/doi/10.1162/tacl_a_00608/2178846/tacl_a_00608.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Michael Wilson","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I32971472","display_name":"Yale University","ror":"https://ror.org/03v76x132","country_code":"US","type":"education","lineage":["https://openalex.org/I32971472"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Michael Wilson","raw_affiliation_strings":["Yale University, USA. michael.a.wilson@yale.edu"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yale University, USA. michael.a.wilson@yale.edu","institution_ids":["https://openalex.org/I32971472"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020894677","display_name":"Jackson Petty","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Jackson Petty","raw_affiliation_strings":["New York University, USA. petty@nyu.edu"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"New York University, USA. petty@nyu.edu","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5071457812","display_name":"Robert Frank","orcid":"https://orcid.org/0000-0001-9253-1658","observed_orcids":["https://orcid.org/0000-0001-9253-1658"]},"institutions":[{"id":"https://openalex.org/I32971472","display_name":"Yale University","ror":"https://ror.org/03v76x132","country_code":"US","type":"education","lineage":["https://openalex.org/I32971472"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Robert Frank","raw_affiliation_strings":["Yale University, USA. bob.frank@yale.edu"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Yale University, USA. bob.frank@yale.edu","institution_ids":["https://openalex.org/I32971472"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5020894677","https://openalex.org/A5071457812"],"corresponding_institution_ids":["https://openalex.org/I32971472","https://openalex.org/I57206974"],"apc_list":null,"apc_paid":null,"fwci":1.222,"has_fulltext":true,"cited_by_count":13,"citation_normalized_percentile":{"value":0.8016907,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":99},"biblio":{"volume":"11","issue":null,"first_page":"1377","last_page":"1395"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.44999998807907104,"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":0.44999998807907104,"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":0.44339999556541443,"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/T10034","display_name":"Syntax, Semantics, Linguistic Variation","score":0.0778999999165535,"subfield":{"id":"https://openalex.org/subfields/1203","display_name":"Language and Linguistics"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.8349999785423279},{"id":"https://openalex.org/keywords/linguistic-generalization","display_name":"linguistic generalization","score":0.8610000014305115},{"id":"https://openalex.org/keywords/argument-structure","display_name":"argument structure","score":0.9950000047683716},{"id":"https://openalex.org/keywords/word-embeddings","display_name":"word embeddings","score":0.4819999933242798},{"id":"https://openalex.org/keywords/passive-voice","display_name":"passive voice","score":0.47699999809265137},{"id":"https://openalex.org/keywords/transformer","display_name":"transformer models","score":0.335999995470047}],"concepts":[{"id":"https://openalex.org/C98184364","wikidata":"https://www.wikidata.org/wiki/Q1780131","display_name":"Argument (complex analysis)","level":2,"score":0.7562944889068604},{"id":"https://openalex.org/C2776397901","wikidata":"https://www.wikidata.org/wiki/Q24905","display_name":"Verb","level":2,"score":0.7179824709892273},{"id":"https://openalex.org/C70777604","wikidata":"https://www.wikidata.org/wiki/Q257885","display_name":"Word order","level":2,"score":0.645224928855896},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6416436433792114},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.6345990300178528},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.6270377039909363},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.5386495590209961},{"id":"https://openalex.org/C121934690","wikidata":"https://www.wikidata.org/wiki/Q1084","display_name":"Noun","level":2,"score":0.49304744601249695},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.49163419008255005},{"id":"https://openalex.org/C2777855551","wikidata":"https://www.wikidata.org/wiki/Q12310021","display_name":"Subject (documents)","level":2,"score":0.4861399829387665},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45018282532691956},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.4106799066066742},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.19192922115325928},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.06401649117469788},{"id":"https://openalex.org/C161191863","wikidata":"https://www.wikidata.org/wiki/Q199655","display_name":"Library science","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},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1162/tacl_a_00608","doi":"https://doi.org/10.1162/tacl_a_00608","is_oa":true,"landing_page_url":"https://doi.org/10.1162/tacl_a_00608","pdf_url":"https://direct.mit.edu/tacl/article-pdf/doi/10.1162/tacl_a_00608/2178846/tacl_a_00608.pdf","source":{"id":"https://openalex.org/S2729999759","display_name":"Transactions of the Association for Computational Linguistics","issn_l":"2307-387X","issn":["2307-387X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["ccf-b","cwts-core","doaj","dongbi-b","jufo-1","norway-2","poland-20","russia-white-list-1"],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Transactions of the Association for Computational Linguistics","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2311.04900","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2311.04900","pdf_url":"https://arxiv.org/pdf/2311.04900","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:doaj.org/article:de75fd3c5eff489a945d1ea1304cfec9","doi":null,"is_oa":true,"landing_page_url":"https://doaj.org/article/de75fd3c5eff489a945d1ea1304cfec9","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Transactions of the Association for Computational Linguistics, Vol 11 (2023)","raw_type":"article"},{"id":"doi:10.48550/arxiv.2311.04900","doi":"https://doi.org/10.48550/arxiv.2311.04900","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2311.04900","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.1162/tacl_a_00608","doi":"https://doi.org/10.1162/tacl_a_00608","is_oa":true,"landing_page_url":"https://doi.org/10.1162/tacl_a_00608","pdf_url":"https://direct.mit.edu/tacl/article-pdf/doi/10.1162/tacl_a_00608/2178846/tacl_a_00608.pdf","source":{"id":"https://openalex.org/S2729999759","display_name":"Transactions of the Association for Computational Linguistics","issn_l":"2307-387X","issn":["2307-387X"],"is_oa":true,"is_in_doaj":true,"is_core":true,"listed_in":["ccf-b","cwts-core","doaj","dongbi-b","jufo-1","norway-2","poland-20","russia-white-list-1"],"host_organization":"https://openalex.org/P4310320244","host_organization_name":"Association for Computational Linguistics","host_organization_lineage":["https://openalex.org/P4310320244"],"host_organization_lineage_names":["Association for Computational Linguistics"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Transactions of the Association for Computational Linguistics","raw_type":"journal-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[{"id":"https://metadata.un.org/sdg/4","score":0.8299999833106995,"display_name":"Quality education"}],"study_designs":[],"awards":[{"id":"https://openalex.org/G2419356789","display_name":"Collaborative Research: Inductive Biases for the Acquisition of Syntactic Transformations in Neural Networks","funder_award_id":"1919321","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320308380","display_name":"Yale University","ror":"https://ror.org/03v76x132"},{"id":"https://openalex.org/F4320309475","display_name":"Northwestern University","ror":"https://ror.org/000e0be47"},{"id":"https://openalex.org/F4320321951","display_name":"Heinrich-Heine-Universit\u00e4t D\u00fcsseldorf","ror":"https://ror.org/024z2rq82"},{"id":"https://openalex.org/F4320322015","display_name":"University of Toronto","ror":"https://ror.org/03dbr7087"},{"id":"https://openalex.org/F4320331904","display_name":"Stony Brook University","ror":"https://ror.org/05qghxh33"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4388778142.pdf","grobid_xml":"https://content.openalex.org/works/W4388778142.grobid-xml"},"referenced_works_count":48,"referenced_works":["https://openalex.org/W1602475114","https://openalex.org/W1682403713","https://openalex.org/W1997161938","https://openalex.org/W2052262800","https://openalex.org/W2060277733","https://openalex.org/W2549835527","https://openalex.org/W2602856279","https://openalex.org/W2728435982","https://openalex.org/W2888922637","https://openalex.org/W2890027146","https://openalex.org/W2908854766","https://openalex.org/W2910243263","https://openalex.org/W2918871811","https://openalex.org/W2946359678","https://openalex.org/W2946417913","https://openalex.org/W2948947170","https://openalex.org/W2962926715","https://openalex.org/W2963341956","https://openalex.org/W2965373594","https://openalex.org/W2965647350","https://openalex.org/W2972752636","https://openalex.org/W2973154008","https://openalex.org/W2973957133","https://openalex.org/W2978017171","https://openalex.org/W2990267819","https://openalex.org/W2996728628","https://openalex.org/W3034199299","https://openalex.org/W3034237830","https://openalex.org/W3034368386","https://openalex.org/W3092478955","https://openalex.org/W3096559634","https://openalex.org/W3116605602","https://openalex.org/W3117575631","https://openalex.org/W3125516434","https://openalex.org/W3130147842","https://openalex.org/W3155746503","https://openalex.org/W3174617925","https://openalex.org/W3196813608","https://openalex.org/W3197876970","https://openalex.org/W3202908475","https://openalex.org/W3208482841","https://openalex.org/W3215599159","https://openalex.org/W4211253404","https://openalex.org/W4224080890","https://openalex.org/W4226052589","https://openalex.org/W4226412047","https://openalex.org/W6766673545","https://openalex.org/W6958741325"],"related_works":["https://openalex.org/W7108476279","https://openalex.org/W4402808117","https://openalex.org/W4412888986","https://openalex.org/W4406231702","https://openalex.org/W4404781505","https://openalex.org/W4400325183","https://openalex.org/W4405173231","https://openalex.org/W4401043619","https://openalex.org/W2991429427","https://openalex.org/W7166499899"],"abstract_inverted_index":{"Abstract":[0],"Language":[1],"models":[2,42,162],"are":[3],"typically":[4],"evaluated":[5],"on":[6,48,151],"their":[7,170],"success":[8],"at":[9,107],"predicting":[10],"the":[11,34,49,62,78,85,93,98,130],"distribution":[12,63],"of":[13,51,64,84,92,97,136],"specific":[14,17],"words":[15],"in":[16,60,141],"contexts.":[18],"Yet":[19],"linguistic":[20],"knowledge":[21],"also":[22],"encodes":[23],"relationships":[24],"between":[25,29,69,109,129],"contexts,":[26],"allowing":[27],"inferences":[28],"word":[30,102],"distributions.":[31],"We":[32,54],"investigate":[33],"degree":[35],"to":[36,148,157,165],"which":[37,120,169],"pre-trained":[38],"transformer-based":[39],"large":[40],"language":[41],"(LLMs)":[43],"represent":[44],"such":[45],"relationships,":[46],"focusing":[47],"domain":[50],"argument":[52,68],"structure.":[53],"find":[55],"that":[56,72,112],"LLMs":[57,105,144],"perform":[58],"well":[59],"generalizing":[61],"a":[65,146,158,166],"novel":[66],"noun":[67],"related":[70,110],"contexts":[71,111],"were":[73],"seen":[74],"during":[75,117],"pre-training":[76],"(e.g.,":[77,128],"active":[79,131],"object":[80,132],"and":[81,133,163],"passive":[82,134],"subject":[83,135],"verb":[86],"spray),":[87],"succeeding":[88],"by":[89],"making":[90],"use":[91],"semantically":[94],"organized":[95],"structure":[96],"embedding":[99],"space":[100],"for":[101,168],"embeddings.":[103],"However,":[104],"fail":[106],"generalizations":[108,127],"have":[113],"not":[114],"been":[115],"observed":[116],"pre-training,":[118],"but":[119,124],"instantiate":[121],"more":[122],"abstract,":[123],"well-attested":[125],"structural":[126],"an":[137],"arbitrary":[138],"verb).":[139],"Instead,":[140],"this":[142],"case,":[143],"show":[145],"bias":[147],"generalize":[149],"based":[150],"linear":[152],"order.":[153],"This":[154],"finding":[155],"points":[156,164],"limitation":[159],"with":[160],"current":[161],"reason":[167],"training":[171],"is":[172],"data-intensive.1":[173]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":8},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":1}],"updated_date":"2026-10-31T12:09:11.096373","created_date":"2023-11-18T00:00:00"}