{"id":"https://openalex.org/W4387947619","doi":"https://doi.org/10.48550/arxiv.2310.16049","title":"MuSR: Testing the Limits of Chain-of-thought with Multistep Soft Reasoning","display_name":"MuSR: Testing the Limits of Chain-of-thought with Multistep Soft Reasoning","publication_year":2023,"publication_date":"2023-10-24","ids":{"openalex":"https://openalex.org/W4387947619","doi":"https://doi.org/10.48550/arxiv.2310.16049"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2310.16049","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.16049","pdf_url":"https://arxiv.org/pdf/2310.16049","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/2310.16049","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5077384760","display_name":"Zayne Sprague","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sprague, Zayne","raw_affiliation_strings":["Department of Computer Science The University of Texas in Austin"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science The University of Texas in Austin","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100355602","display_name":"Xi Ye","orcid":"https://orcid.org/0000-0003-0331-7946","observed_orcids":["https://orcid.org/0000-0003-0331-7946"]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ye, Xi","raw_affiliation_strings":["Department of Computer Science The University of Texas in Austin"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science The University of Texas in Austin","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050663933","display_name":"Kaj Bostrom","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bostrom, Kaj","raw_affiliation_strings":["Department of Computer Science The University of Texas in Austin"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science The University of Texas in Austin","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057341982","display_name":"Swarat Chaudhuri","orcid":"https://orcid.org/0000-0002-6859-1391","observed_orcids":["https://orcid.org/0000-0002-6859-1391"]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chaudhuri, Swarat","raw_affiliation_strings":["Department of Computer Science The University of Texas in Austin"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science The University of Texas in Austin","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5015133105","display_name":"Greg Durrett","orcid":"https://orcid.org/0000-0002-7061-7298","observed_orcids":["https://orcid.org/0000-0002-7061-7298"]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Durrett, Greg","raw_affiliation_strings":["Department of Computer Science The University of Texas in Austin"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science The University of Texas in Austin","institution_ids":["https://openalex.org/I86519309"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I86519309"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":6,"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":0.34940001368522644,"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":0.34940001368522644,"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/T13643","display_name":"Artificial Intelligence in Law","score":0.13130000233650208,"subfield":{"id":"https://openalex.org/subfields/3320","display_name":"Political Science and International Relations"},"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/T13904","display_name":"Artificial Intelligence Applications","score":0.12520000338554382,"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/large-language-models","display_name":"large language models","score":0.984000027179718},{"id":"https://openalex.org/keywords/crime-scene-investigation","display_name":"crime scene investigation","score":0.9580000042915344},{"id":"https://openalex.org/keywords/crime-scene-reconstruction","display_name":"crime scene reconstruction","score":0.8230000138282776},{"id":"https://openalex.org/keywords/multi-step-reasoning","display_name":"multi-step reasoning","score":0.7110000252723694},{"id":"https://openalex.org/keywords/abstention","display_name":"abstention","score":0.675000011920929},{"id":"https://openalex.org/keywords/murder-mystery","display_name":"murder mystery","score":0.6679999828338623},{"id":"https://openalex.org/keywords/object-placement","display_name":"object placement","score":0.8980000019073486},{"id":"https://openalex.org/keywords/claude","display_name":"Claude","score":0.5899999737739563},{"id":"https://openalex.org/keywords/qwen","display_name":"Qwen","score":0.9599999785423279},{"id":"https://openalex.org/keywords/deepseek","display_name":"DeepSeek","score":0.9670000076293945},{"id":"https://openalex.org/keywords/gemma","display_name":"Gemma","score":0.6629999876022339},{"id":"https://openalex.org/keywords/token-cost","display_name":"token cost","score":0.3880000114440918},{"id":"https://openalex.org/keywords/prompt-engineering","display_name":"prompt engineering","score":0.39500001072883606},{"id":"https://openalex.org/keywords/chain-of-thought-prompting","display_name":"chain-of-thought prompting","score":0.9210000038146973},{"id":"https://openalex.org/keywords/llm-performance-comparison","display_name":"LLM performance comparison","score":0.34299999475479126},{"id":"https://openalex.org/keywords/llm-hallucination","display_name":"LLM hallucination","score":0.3400000035762787}],"concepts":[{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.7449340224266052},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6897574663162231},{"id":"https://openalex.org/C199033989","wikidata":"https://www.wikidata.org/wiki/Q1318295","display_name":"Narrative","level":2,"score":0.6842260956764221},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46090492606163025},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.4461812674999237},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.4226968586444855},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.37139421701431274},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.07497745752334595},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2310.16049","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.16049","pdf_url":"https://arxiv.org/pdf/2310.16049","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.2310.16049","doi":"https://doi.org/10.48550/arxiv.2310.16049","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2310.16049","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":"article-journal"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2310.16049","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.16049","pdf_url":"https://arxiv.org/pdf/2310.16049","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":[],"sustainable_development_goals_aurora":[{"id":"https://metadata.un.org/sdg/16","score":0.6000000238418579,"display_name":"Peace, justice, and strong institutions"}],"study_designs":[],"awards":[{"id":"https://openalex.org/G4522540643","display_name":"CAREER: Flexible and Robust Reasoning in Natural Language","funder_award_id":"2145280","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4713059963","display_name":null,"funder_award_id":"FA8750","funder_id":"https://openalex.org/F4320332180","funder_display_name":"Defense Advanced Research Projects Agency"},{"id":"https://openalex.org/G6671297155","display_name":null,"funder_award_id":"CAREER","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/F4320315389","display_name":"Open Philanthropy Project","ror":"https://ror.org/004d1k391"},{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"},{"id":"https://openalex.org/F4320338294","display_name":"Air Force Research Laboratory","ror":"https://ror.org/02e2egq70"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4387947619.pdf","grobid_xml":"https://content.openalex.org/works/W4387947619.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7214371292","https://openalex.org/W7154765720","https://openalex.org/W7093361127","https://openalex.org/W4403286340","https://openalex.org/W4412075512","https://openalex.org/W4385718074","https://openalex.org/W4413328395","https://openalex.org/W7166808773","https://openalex.org/W7084035800","https://openalex.org/W7117415714"],"abstract_inverted_index":{"While":[0],"large":[1],"language":[2,58,69],"models":[3,59],"(LLMs)":[4],"equipped":[5],"with":[6,153],"techniques":[7,164,175],"like":[8,45,176],"chain-of-thought":[9,177],"prompting":[10,163],"have":[11,48],"demonstrated":[12],"impressive":[13],"capabilities,":[14],"they":[15],"still":[16],"fall":[17],"short":[18],"in":[19,25,66,104],"their":[20],"ability":[21],"to":[22,38,127,151,178],"reason":[23],"robustly":[24],"complex":[26,92],"settings.":[27],"However,":[28],"evaluating":[29,57],"LLM":[30],"reasoning":[31,63,93],"is":[32,79],"challenging":[33,138],"because":[34],"system":[35],"capabilities":[36],"continue":[37],"grow":[39],"while":[40,143],"benchmark":[41],"datasets":[42],"for":[43,56,148,174],"tasks":[44,64],"logical":[46],"deduction":[47],"remained":[49],"static.":[50],"We":[51,156],"introduce":[52],"MuSR,":[53],"a":[54,67,82,158],"dataset":[55,72,120,167],"on":[60,165],"multistep":[61],"soft":[62],"specified":[65],"natural":[68],"narrative.":[70],"This":[71],"has":[73],"two":[74],"crucial":[75],"features.":[76],"First,":[77],"it":[78,134],"created":[80],"through":[81],"novel":[83],"neurosymbolic":[84],"synthetic-to-natural":[85],"generation":[86],"algorithm,":[87],"enabling":[88],"the":[89,170],"construction":[90],"of":[91,130,160],"instances":[94,121],"that":[95,172],"challenge":[96],"GPT-4":[97],"(e.g.,":[98],"murder":[99],"mysteries":[100],"roughly":[101],"1000":[102],"words":[103],"length)":[105],"and":[106,146,162,168],"which":[107],"can":[108],"be":[109],"scaled":[110],"further":[111],"as":[112],"more":[113,137],"capable":[114],"LLMs":[115,161],"are":[116,122],"released.":[117],"Second,":[118],"our":[119],"free":[123],"text":[124],"narratives":[125],"corresponding":[126],"real-world":[128],"domains":[129],"reasoning;":[131],"this":[132,166],"makes":[133],"simultaneously":[135],"much":[136],"than":[139],"other":[140],"synthetically-crafted":[141],"benchmarks":[142],"remaining":[144],"realistic":[145],"tractable":[147],"human":[149],"annotators":[150],"solve":[152],"high":[154],"accuracy.":[155],"evaluate":[157],"range":[159],"characterize":[169],"gaps":[171],"remain":[173],"perform":[179],"robust":[180],"reasoning.":[181]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":1}],"updated_date":"2026-10-29T09:56:32.521200","created_date":"2023-10-26T00:00:00"}