{"id":"https://openalex.org/W7204451494","doi":"https://doi.org/10.48550/arxiv.2608.25005","title":"The Imperfective Paradox Is Not Necessarily in Large Language Models: A Benchmark Failure Before a Model Failure","display_name":"The Imperfective Paradox Is Not Necessarily in Large Language Models: A Benchmark Failure Before a Model Failure","publication_year":2026,"publication_date":"2026-08-25","ids":{"openalex":"https://openalex.org/W7204451494","doi":"https://doi.org/10.48550/arxiv.2608.25005"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.25005","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.25005","pdf_url":"https://arxiv.org/pdf/2608.25005","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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.25005","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137009268","display_name":"Kaiqiao Han","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Han, Kaiqiao","raw_affiliation_strings":["University of California , Los Angeles"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California , Los Angeles","institution_ids":["https://openalex.org/I161318765"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5126252841","display_name":"Yizhou Sun","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sun, Yizhou","raw_affiliation_strings":["University of California , Los Angeles"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California , Los Angeles","institution_ids":["https://openalex.org/I161318765"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I161318765"],"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":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.4666000008583069,"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":0.4666000008583069,"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":0.2485000044107437,"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"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.24130000174045563,"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/culmination","display_name":"culmination","score":0.5590000152587891},{"id":"https://openalex.org/keywords/natural-language-inference","display_name":"natural language inference","score":0.6859999895095825},{"id":"https://openalex.org/keywords/compositional-semantics","display_name":"compositional semantics","score":0.9089999794960022},{"id":"https://openalex.org/keywords/aspectual-classification","display_name":"aspectual classification","score":0.6549999713897705},{"id":"https://openalex.org/keywords/minimal-pairs","display_name":"minimal pairs","score":0.6140000224113464},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.6890000104904175},{"id":"https://openalex.org/keywords/qwen","display_name":"Qwen","score":0.5809999704360962},{"id":"https://openalex.org/keywords/gpt-5-4","display_name":"GPT-5.4","score":0.7900000214576721}],"concepts":[{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.7566999793052673},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6625999808311462},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6326000094413757},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5376999974250793},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4489000141620636},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.42260000109672546},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.4180000126361847},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.32089999318122864},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.31859999895095825},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3151000142097473},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.2939000129699707},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.29030001163482666},{"id":"https://openalex.org/C2776818064","wikidata":"https://www.wikidata.org/wiki/Q829903","display_name":"Agreement","level":2,"score":0.27810001373291016},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.27160000801086426},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.2678999900817871}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.25005","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.25005","pdf_url":"https://arxiv.org/pdf/2608.25005","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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.25005","doi":"https://doi.org/10.48550/arxiv.2608.25005","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.25005","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.25005","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.25005","pdf_url":"https://arxiv.org/pdf/2608.25005","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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204451494.pdf","grobid_xml":"https://content.openalex.org/works/W7204451494.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7166873957","https://openalex.org/W7166863632","https://openalex.org/W7160038594","https://openalex.org/W2800397671","https://openalex.org/W7120064106","https://openalex.org/W7205426391","https://openalex.org/W4411630266","https://openalex.org/W4402683412","https://openalex.org/W4412945525","https://openalex.org/W7140138029"],"abstract_inverted_index":{"The":[0],"imperfective":[1],"paradox":[2],"provides":[3],"a":[4,33,43,82,142,172,186],"useful":[5],"test":[6],"of":[7,87,102,108,237,251],"compositional":[8,211],"semantic":[9,150,196],"analysis.":[10],"Recent":[11],"work":[12],"constructs":[13],"an":[14,117],"NLI":[15,84,140],"benchmark":[16,49,75],"and":[17,50,52,61,79,106,124,146,152,198,201,214,228,240],"reports":[18],"that":[19,39,54,158,182,242,250],"models":[20,159,244],"frequently":[21],"infer":[22],"completed":[23],"telic":[24],"events":[25],"from":[26],"progressive":[27],"descriptions,":[28],"attributing":[29],"this":[30],"behavior":[31],"to":[32,115,249],"Teleological":[34],"Bias.":[35,178],"It":[36],"further":[37,180],"argues":[38],"prompting":[40,183],"interventions":[41,184],"cause":[42],"Calibration":[44],"Crisis.":[45],"We":[46,64,179],"reexamine":[47],"the":[48,74,109,134,168,194,234],"conclusions":[51],"show":[53,157,181],"it":[55],"is":[56],"substantially":[57],"affected":[58],"by":[59],"conceptual":[60,67],"evaluation":[62,135],"mis-specifications.":[63,68],"identify":[65,204],"three":[66],"In":[69,97],"particular,":[70],"Aspectual":[71],"Reduction":[72],"affects":[73],"construction,":[76],"analysis,":[77],"experiments,":[78],"conclusions.":[80],"Under":[81],"strict":[83],"standard,":[85],"76%":[86],"Group":[88,103,110],"A":[89,104],"instances":[90],"do":[91,161],"not":[92,162],"explicitly":[93],"rule":[94],"out":[95],"culmination.":[96],"our":[98],"native-speaker":[99],"annotation,":[100],"38%":[101],"examples":[105,112],"29%":[107],"C":[111],"were":[113],"judged":[114],"permit":[116],"alternative":[118],"interpretation.":[119],"To":[120],"control":[121],"these":[122,243],"issues":[123],"lexical":[125],"variation,":[126],"we":[127,137,174],"construct":[128],"Lexically":[129],"Matched":[130],"Minimal":[131],"Pairs.":[132],"At":[133],"level,":[136],"formulate":[138],"event-semantic":[139],"as":[141,176],"Multi-step":[143],"Reasoning":[144],"Problem":[145],"assess":[147],"both":[148],"intermediate":[149],"decisions":[151],"final":[153],"predictions.":[154],"Our":[155,220],"results":[156],"often":[160],"affirm":[163],"culmination":[164],"but":[165],"nevertheless":[166],"accept":[167],"corresponding":[169],"simple-past":[170],"hypothesis,":[171],"pattern":[173],"characterize":[175],"Sufficiency":[177],"produce":[185],"Decision":[187],"Shift":[188],"among":[189],"labels":[190],"without":[191],"reliably":[192],"improving":[193],"underlying":[195],"understanding":[197],"reasoning.":[199],"Intermediate":[200],"oracle-guided":[202],"analyses":[203],"two":[205],"additional":[206],"failure":[207],"modes:":[208],"errors":[209],"in":[210],"aspectual":[212,238],"classification":[213,239],"Surface-form":[215],"Attraction":[216],"toward":[217],"surface-associated":[218],"answers.":[219],"experiments":[221],"on":[222],"Qwen-7B":[223],"with":[224],"suitable":[225],"prompts,":[226],"GPT-5.4,":[227],"Qwen-72B":[229],"provide":[230],"initial":[231],"evidence":[232],"for":[233],"context":[235],"sensitivity":[236],"suggest":[241],"can":[245],"achieve":[246],"performance":[247],"comparable":[248],"human":[252],"annotators.":[253]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-08-28T00:00:00"}