{"id":"https://openalex.org/W4412945597","doi":"https://doi.org/10.18653/v1/2025.acl-long.433","title":"Large Language and Reasoning Models are Shallow Disjunctive Reasoners","display_name":"Large Language and Reasoning Models are Shallow Disjunctive Reasoners","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412945597","doi":"https://doi.org/10.18653/v1/2025.acl-long.433"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.433","is_oa":false,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.433","pdf_url":null,"source":null,"license":null,"license_id":null,"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":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://orca.cardiff.ac.uk/id/eprint/179567/1/_ARR_February_2025__Can_large_language_and_reasoning_models_do_systematic_relational_reasoning_.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5077247933","display_name":"Irtaza Khalid","orcid":"https://orcid.org/0000-0002-1532-7829"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Irtaza Khalid","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5093622210","display_name":"Amir Masoud Nourollah","orcid":"https://orcid.org/0000-0002-6083-6673"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Amir Masoud Nourollah","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5052196540","display_name":"Steven Schockaert","orcid":"https://orcid.org/0000-0002-9256-2881"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Steven Schockaert","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":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.15377769,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"8843","last_page":"8869"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.633400022983551,"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":0.633400022983551,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.6266999840736389,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.5690000057220459,"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":0.7542523145675659},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.41854187846183777},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3981212079524994},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.39411988854408264},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.32029351592063904}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7542523145675659},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.41854187846183777},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3981212079524994},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.39411988854408264},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.32029351592063904}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.433","is_oa":false,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.433","pdf_url":null,"source":null,"license":null,"license_id":null,"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"},{"id":"pmh:oai:https://orca.cardiff.ac.uk:179567","is_oa":true,"landing_page_url":null,"pdf_url":"https://orca.cardiff.ac.uk/id/eprint/179567/1/_ARR_February_2025__Can_large_language_and_reasoning_models_do_systematic_relational_reasoning_.pdf","source":{"id":"https://openalex.org/S4306401195","display_name":"ORCA Online Research @Cardiff (Cardiff University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I79510175","host_organization_name":"Cardiff University","host_organization_lineage":["https://openalex.org/I79510175"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"PeerReviewed"}],"best_oa_location":{"id":"pmh:oai:https://orca.cardiff.ac.uk:179567","is_oa":true,"landing_page_url":null,"pdf_url":"https://orca.cardiff.ac.uk/id/eprint/179567/1/_ARR_February_2025__Can_large_language_and_reasoning_models_do_systematic_relational_reasoning_.pdf","source":{"id":"https://openalex.org/S4306401195","display_name":"ORCA Online Research @Cardiff (Cardiff University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I79510175","host_organization_name":"Cardiff University","host_organization_lineage":["https://openalex.org/I79510175"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"PeerReviewed"},"sustainable_development_goals":[{"display_name":"Quality Education","score":0.5799999833106995,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4412945597.pdf"},"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":{"Large":[0,7],"Language":[1,8],"Models":[2,9],"(LLMs)":[3,10],"and":[4,53,80,106],"Systematic":[5],"Reasoning":[6,75],"have":[11,56],"been":[12,58],"found":[13],"to":[14,25,41,117],"struggle":[15,137],"with":[16],"systematic":[17,100],"reasoning.":[18,108],"Even":[19],"on":[20,32,43,50,96],"tasks":[21,97,135],"where":[22,84],"they":[23],"appear":[24],"perform":[26],"well,":[27],"their":[28,129],"performance":[29],"often":[30],"depends":[31],"shortcuts":[33],"rather":[34],"than":[35],"genuine":[36,85],"reasoning":[37,134],"abilities,":[38],"leading":[39],"them":[40],"collapse":[42],"out-of-distribution":[44],"(OOD)":[45],"examples.":[46],"Post-training":[47],"strategies":[48],"based":[49],"reinforcement":[51],"learning":[52],"chain-of-thought":[54],"prompting":[55],"recently":[57],"hailed":[59],"as":[60],"a":[61],"step":[62],"change.":[63],"However,":[64],"little":[65],"is":[66],"known":[67],"about":[68],"the":[69,72,139,161],"potential":[70],"of":[71,153,164],"resulting":[73],"\u201cLarge":[74],"Models\u201d":[76],"(LRMs)":[77],"beyond":[78],"maths":[79],"programming-based":[81],"problem":[82,115],"solving,":[83],"OOD":[86,120],"problems":[87],"can":[88],"be":[89],"sparse.":[90],"In":[91],"this":[92],"paper,":[93],"we":[94],"focus":[95],"that":[98,124,166],"require":[99],"relational":[101],"composition":[102],"for":[103,160],"qualitative":[104],"spatial":[105],"temporal":[107],"The":[109],"setting":[110],"allows":[111],"fine":[112],"control":[113],"over":[114],"difficulty":[116],"precisely":[118],"measure":[119],"generalization.":[121,155],"We":[122,156],"find":[123],"zero-shot":[125],"LRMs":[126,167],"generally":[127],"outperform":[128],"LLM":[130],"counterparts":[131],"in":[132,138],"single-path":[133],"but":[136],"multi-path":[140,154],"setting.":[141],"While":[142],"showing":[143],"comparatively":[144],"better":[145],"results,":[146],"fine-tuned":[147],"LLMs":[148],"are":[149,168],"also":[150,157],"not":[151],"capable":[152],"provide":[158],"evidence":[159],"behavioral":[162],"interpretation":[163],"this\u2014namely,":[165],"shallow":[169],"disjunctive":[170],"reasoners.":[171]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}