{"id":"https://openalex.org/W7140092266","doi":"https://doi.org/10.18653/v1/2026.eacl-long.19","title":"The Dog the Cat Chased Stumped the Model: Measuring When Language Models Abandon Structure for Shortcuts","display_name":"The Dog the Cat Chased Stumped the Model: Measuring When Language Models Abandon Structure for Shortcuts","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7140092266","doi":"https://doi.org/10.18653/v1/2026.eacl-long.19"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.eacl-long.19","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.19","pdf_url":"https://aclanthology.org/2026.eacl-long.19.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 19th Conference of the European Chapter 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/2026.eacl-long.19.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130332467","display_name":"Sangmitra Madhusudan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sangmitra Madhusudan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130332390","display_name":"Kaige Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kaige Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130373428","display_name":"Ali Emami","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ali Emami","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"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":0.39074314,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"428","last_page":"453"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12090","display_name":"Language and cultural evolution","score":0.23729999363422394,"subfield":{"id":"https://openalex.org/subfields/3316","display_name":"Cultural Studies"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12090","display_name":"Language and cultural evolution","score":0.23729999363422394,"subfield":{"id":"https://openalex.org/subfields/3316","display_name":"Cultural Studies"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T13910","display_name":"Computational and Text Analysis Methods","score":0.15880000591278076,"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"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.06729999929666519,"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/language-model","display_name":"Language model","score":0.3147999942302704},{"id":"https://openalex.org/keywords/modeling-language","display_name":"Modeling language","score":0.3147999942302704},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.27480000257492065},{"id":"https://openalex.org/keywords/mathematical-model","display_name":"Mathematical model","score":0.2556999921798706}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5403000116348267},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.31779998540878296},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3147999942302704},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.3147999942302704},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.27480000257492065},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.2653999924659729},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.2581999897956848},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.25600001215934753},{"id":"https://openalex.org/C76969082","wikidata":"https://www.wikidata.org/wiki/Q486902","display_name":"Mathematical model","level":2,"score":0.2556999921798706},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2379000037908554}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.eacl-long.19","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.19","pdf_url":"https://aclanthology.org/2026.eacl-long.19.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 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.eacl-long.19","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.eacl-long.19","pdf_url":"https://aclanthology.org/2026.eacl-long.19.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 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7140092266.pdf","grobid_xml":"https://content.openalex.org/works/W7140092266.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"When":[0],"language":[1],"models":[2,98,113,149,162,182],"correctly":[3],"parse":[4],"\"The":[5,48],"cat":[6,49],"that":[7,100],"the":[8,51,176],"dog":[9,52],"chased":[10],"meowed,\"":[11],"are":[12],"they":[13,124],"analyzing":[14],"syntax":[15],"or":[16],"simply":[17],"familiar":[18],"with":[19,110,112,168],"dogs":[20],"chasing":[21],"cats?Despite":[22],"extensive":[23],"benchmarking,":[24],"we":[25],"lack":[26],"methods":[27],"to":[28,65,118,179,187],"distinguish":[29],"structural":[30,126,185],"understanding":[31],"from":[32,63,184],"semantic":[33,129,131,147,156,173],"pattern":[34,188],"matching.We":[35],"introduce":[36],"CENTERBENCH,":[37],"a":[38,71],"dataset":[39],"of":[40],"9,720":[41],"comprehension":[42,87],"questions":[43,88,136],"on":[44,135],"centerembedded":[45],"sentences":[46,107],"(like":[47],"[that":[50],"chased]":[53],"meowed\")":[54],"where":[55,140],"relative":[56],"clauses":[57],"nest":[58],"recursively,":[59],"creating":[60],"processing":[61],"demands":[62],"simple":[64],"deeply":[66],"nested":[67],"structures.Each":[68],"sentence":[69],"has":[70],"syntactically":[72],"identical":[73],"but":[74,152],"semantically":[75],"implausible":[76,106],"counterpart":[77],"(e.g.,":[78],"mailmen":[79],"prescribe":[80],"medicine,":[81],"doctors":[82],"deliver":[83],"mail)":[84],"and":[85,94,105,159],"six":[86,97],"testing":[89],"surface":[90],"understanding,":[91],"syntactic":[92],"dependencies,":[93],"causal":[95,142],"reasoning.Testing":[96],"reveals":[99],"performance":[101,134],"gaps":[102,116],"between":[103],"plausible":[104],"widen":[108],"systematically":[109,166],"complexity,":[111,169],"showing":[114],"median":[115],"up":[117],"26.8":[119],"percentage":[120],"points,":[121],"quantifying":[122],"when":[123,181],"abandon":[125],"analysis":[127,186],"for":[128],"associations.Notably,":[130],"plausibility":[132,164],"harms":[133],"about":[137],"resulting":[138],"actions,":[139],"following":[141],"relationships":[143],"matters":[144],"more":[145],"than":[146],"coherence.Reasoning":[148],"improve":[150],"accuracy":[151],"their":[153],"traces":[154],"show":[155],"shortcuts,":[157],"overthinking,":[158],"answer":[160],"refusal.Unlike":[161],"whose":[163],"advantage":[165],"widens":[167],"humans":[170],"shows":[171],"variable":[172],"effects.CENTERBENCH":[174],"provides":[175],"first":[177],"framework":[178],"identify":[180],"shift":[183],"matching.":[189]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-03-24T00:00:00"}