{"id":"https://openalex.org/W4399401573","doi":"https://doi.org/10.48550/arxiv.2406.02061","title":"Alice in Wonderland: Simple Tasks Showing Complete Reasoning Breakdown in State-Of-the-Art Large Language Models","display_name":"Alice in Wonderland: Simple Tasks Showing Complete Reasoning Breakdown in State-Of-the-Art Large Language Models","publication_year":2024,"publication_date":"2024-06-04","ids":{"openalex":"https://openalex.org/W4399401573","doi":"https://doi.org/10.48550/arxiv.2406.02061"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2406.02061","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2406.02061","pdf_url":"https://arxiv.org/pdf/2406.02061","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"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2406.02061","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5042252873","display_name":"Marianna Nezhurina","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nezhurina, Marianna","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059737073","display_name":"Lucia Cipolina-Kun","orcid":"https://orcid.org/0000-0003-3094-577X","observed_orcids":["https://orcid.org/0000-0003-3094-577X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cipolina-Kun, Lucia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050394079","display_name":"Mehdi Cherti","orcid":"https://orcid.org/0000-0001-5865-0469","observed_orcids":["https://orcid.org/0000-0001-5865-0469"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cherti, Mehdi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5085870953","display_name":"Jenia Jitsev","orcid":"https://orcid.org/0000-0002-1221-7851","observed_orcids":["https://orcid.org/0000-0002-1221-7851"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jitsev, Jenia","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":null,"has_fulltext":false,"cited_by_count":22,"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/T10028","display_name":"Topic Modeling","score":0.8765000104904175,"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":0.8765000104904175,"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":0.06340000033378601,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.025100000202655792,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/large-language-models","display_name":"large language models","score":0.9359999895095825},{"id":"https://openalex.org/keywords/commonsense-reasoning","display_name":"commonsense reasoning","score":0.902999997138977},{"id":"https://openalex.org/keywords/overconfidence","display_name":"overconfidence","score":0.8859999775886536},{"id":"https://openalex.org/keywords/confabulation","display_name":"confabulation","score":0.8930000066757202},{"id":"https://openalex.org/keywords/explanation-generation","display_name":"explanation generation","score":0.5659999847412109},{"id":"https://openalex.org/keywords/benchmarking","display_name":"benchmarking","score":0.3659999966621399},{"id":"https://openalex.org/keywords/foundation-models","display_name":"foundation models","score":0.5529999732971191}],"concepts":[{"id":"https://openalex.org/C2778222013","wikidata":"https://www.wikidata.org/wiki/Q1136926","display_name":"Alice (programming language)","level":2,"score":0.9111138582229614},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.7297184467315674},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.5597718954086304},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5374385118484497},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.35457733273506165},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.22517526149749756},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.21901416778564453},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.1292799711227417},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.1137150228023529}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2406.02061","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2406.02061","pdf_url":"https://arxiv.org/pdf/2406.02061","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":"doi:10.48550/arxiv.2406.02061","doi":"https://doi.org/10.48550/arxiv.2406.02061","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2406.02061","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":"pmh:oai:arXiv.org:2406.02061","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2406.02061","pdf_url":"https://arxiv.org/pdf/2406.02061","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"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7138290724","https://openalex.org/W4416284248","https://openalex.org/W4412888699","https://openalex.org/W4416887915","https://openalex.org/W7161231178","https://openalex.org/W7155576675","https://openalex.org/W7151829257","https://openalex.org/W7117854721","https://openalex.org/W7133231369","https://openalex.org/W7213624158"],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"are":[4],"often":[5,139],"described":[6],"as":[7,115],"being":[8],"instances":[9],"of":[10,45,59,74,79,161,194,198,201,218],"foundation":[11],"models":[12,16,81,116,167],"-":[13],"that":[14,17,34,125,213,223],"is,":[15],"transfer":[18],"strongly":[19],"across":[20,56,120],"various":[21,57,159],"tasks":[22,51],"and":[23,50,76,187,234,243],"conditions":[24],"in":[25,47,103,135,150,240],"few-show":[26],"or":[27,164],"zero-shot":[28],"manner,":[29],"while":[30],"exhibiting":[31],"scaling":[32],"laws":[33],"predict":[35],"function":[36,75],"improvement":[37],"when":[38],"increasing":[39],"the":[40,84,136,155,166,170,185,195,241],"pre-training":[41],"scale.":[42],"These":[43],"claims":[44],"excelling":[46],"different":[48],"functions":[49],"rely":[52],"on":[53],"measurements":[54],"taken":[55],"sets":[58],"standardized":[60,211],"benchmarks":[61,212],"showing":[62],"high":[63],"scores":[64],"for":[65,237],"such":[66,219],"models.":[67],"We":[68,179],"demonstrate":[69],"here":[70],"a":[71,93],"dramatic":[72],"breakdown":[73,112],"reasoning":[77,221],"capabilities":[78,197],"state-of-the-art":[80,231],"trained":[82],"at":[83,250],"largest":[85],"available":[86],"scales":[87],"which":[88],"claim":[89],"strong":[90,118,133],"function,":[91],"using":[92],"simple,":[94],"short,":[95],"conventional":[96],"common":[97,207],"sense":[98],"problem":[99,123,129],"(AIW":[100],"problem)":[101],"formulated":[102],"concise":[104],"natural":[105],"language,":[106],"easily":[107],"solvable":[108],"by":[109,142,174,229],"humans.":[110],"The":[111],"is":[113],"dramatic,":[114],"show":[117],"fluctuations":[119],"even":[121],"slight":[122],"variations":[124],"should":[126],"not":[127],"affect":[128],"solving,":[130],"also":[131,205],"expressing":[132],"overconfidence":[134],"wrong":[137,171],"solutions,":[138],"backed":[140],"up":[141],"plausible":[143],"sounding":[144],"explanation-like":[145],"confabulations.":[146],"Various":[147],"standard":[148],"interventions":[149],"an":[151],"attempt":[152],"to":[153,168,184,190,209,226],"get":[154],"right":[156],"solution,":[157],"like":[158],"type":[160],"enhanced":[162],"prompting,":[163],"urging":[165],"reconsider":[169],"solutions":[172],"again":[173],"multi":[175],"step":[176],"re-evaluation,":[177],"fail.":[178],"take":[180],"these":[181],"initial":[182],"observations":[183],"scientific":[186],"technological":[188],"community":[189],"stimulate":[191],"urgent":[192],"re-assessment":[193,204],"claimed":[196],"current":[199,230],"generation":[200],"LLMs.":[202],"Such":[203],"requires":[206],"action":[208],"create":[210],"would":[214],"allow":[215],"proper":[216],"detection":[217],"basic":[220],"deficits":[222],"obviously":[224],"manage":[225],"remain":[227],"undiscovered":[228],"evaluation":[232],"procedures":[233],"benchmarks.":[235],"Code":[236],"reproducing":[238],"experiments":[239,245],"paper":[242],"raw":[244],"data":[246],"can":[247],"be":[248],"found":[249],"https://github.com/LAION-AI/AIW":[251]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":7}],"updated_date":"2026-11-03T10:50:19.383937","created_date":"2024-06-07T00:00:00"}