{"id":"https://openalex.org/W4387724709","doi":"https://doi.org/10.48550/arxiv.2310.10080","title":"Let's reward step by step: Step-Level reward model as the Navigators for Reasoning","display_name":"Let's reward step by step: Step-Level reward model as the Navigators for Reasoning","publication_year":2023,"publication_date":"2023-10-16","ids":{"openalex":"https://openalex.org/W4387724709","doi":"https://doi.org/10.48550/arxiv.2310.10080"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2310.10080","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.10080","pdf_url":"https://arxiv.org/pdf/2310.10080","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-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","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/2310.10080","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Ma, Qianli","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Ma, Qianli","raw_affiliation_strings":["School of Computing , National University of Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computing , National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100533445","display_name":"Haotian Zhou","orcid":"https://orcid.org/0009-0008-7598-010X","observed_orcids":["https://orcid.org/0009-0008-7598-010X","https://orcid.org/0009-0006-5395-0626"]},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Zhou, Haotian","raw_affiliation_strings":["School of Computing , National University of Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computing , National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066756258","display_name":"Tingkai Liu","orcid":"https://orcid.org/0000-0003-3075-7648","observed_orcids":["https://orcid.org/0000-0003-3075-7648"]},"institutions":[{"id":"https://openalex.org/I4405256866","display_name":"ByteDance (China)","ror":"https://ror.org/04m8av638","country_code":"CN","type":"company","lineage":["https://openalex.org/I4405256866"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liu, Tingkai","raw_affiliation_strings":["ByteDance , Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ByteDance , Inc","institution_ids":["https://openalex.org/I4405256866"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004652202","display_name":"Jianbo Yuan","orcid":"https://orcid.org/0000-0001-7949-9841","observed_orcids":["https://orcid.org/0000-0001-7949-9841"]},"institutions":[{"id":"https://openalex.org/I4405256866","display_name":"ByteDance (China)","ror":"https://ror.org/04m8av638","country_code":"CN","type":"company","lineage":["https://openalex.org/I4405256866"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuan, Jianbo","raw_affiliation_strings":["ByteDance , Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ByteDance , Inc","institution_ids":["https://openalex.org/I4405256866"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100355025","display_name":"Pengfei Liu","orcid":"https://orcid.org/0009-0008-6932-7091","observed_orcids":["https://orcid.org/0009-0008-6932-7091"]},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liu, Pengfei","raw_affiliation_strings":["Generative Artificial Intelligence Research Lab , Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Generative Artificial Intelligence Research Lab , Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":null,"display_name":"You, Yang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"You, Yang","raw_affiliation_strings":["School of Computing , National University of Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computing , National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5082599714","display_name":"Hongxia Yang","orcid":"https://orcid.org/0000-0002-0580-9728","observed_orcids":["https://orcid.org/0000-0002-0580-9728"]},"institutions":[{"id":"https://openalex.org/I4405256866","display_name":"ByteDance (China)","ror":"https://ror.org/04m8av638","country_code":"CN","type":"company","lineage":["https://openalex.org/I4405256866"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yang, Hongxia","raw_affiliation_strings":["ByteDance , Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ByteDance , Inc","institution_ids":["https://openalex.org/I4405256866"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"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/T10028","display_name":"Topic Modeling","score":0.7875999808311462,"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":0.7875999808311462,"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":0.08560000360012054,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.019600000232458115,"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/multi-step-reasoning","display_name":"multi-step reasoning","score":0.9700000286102295},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.7379999756813049},{"id":"https://openalex.org/keywords/greedy-search","display_name":"greedy search","score":0.7080000042915344},{"id":"https://openalex.org/keywords/chain-of-thought","display_name":"chain-of-thought","score":0.5649999976158142},{"id":"https://openalex.org/keywords/mathematical-reasoning","display_name":"mathematical reasoning","score":0.9819999933242798},{"id":"https://openalex.org/keywords/code-generation","display_name":"code generation","score":0.9539999961853027},{"id":"https://openalex.org/keywords/gsm8k","display_name":"GSM8K","score":0.7910000085830688},{"id":"https://openalex.org/keywords/math","display_name":"MATH","score":0.9879999756813049}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7898997068405151},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6826990842819214},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.6129752397537231},{"id":"https://openalex.org/C2778061373","wikidata":"https://www.wikidata.org/wiki/Q1315146","display_name":"Predictive coding","level":3,"score":0.526888370513916},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5168910026550293},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5102429389953613},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5086632370948792},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.5016775131225586},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.4482731819152832},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.11530691385269165},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.10522988438606262},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.0},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2310.10080","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.10080","pdf_url":"https://arxiv.org/pdf/2310.10080","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-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2310.10080","doi":"https://doi.org/10.48550/arxiv.2310.10080","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2310.10080","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:2310.10080","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.10080","pdf_url":"https://arxiv.org/pdf/2310.10080","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-sa","license_id":"https://openalex.org/licenses/cc-by-nc-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320322724","display_name":"Ministry of Education, India","ror":"https://ror.org/048xjjh50"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4387724709.pdf","grobid_xml":"https://content.openalex.org/works/W4387724709.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4414602045","https://openalex.org/W4407806651","https://openalex.org/W4416034198","https://openalex.org/W7138854239","https://openalex.org/W4415915064","https://openalex.org/W7092285489","https://openalex.org/W6948132249","https://openalex.org/W4404781622","https://openalex.org/W7133217386","https://openalex.org/W7147465866"],"abstract_inverted_index":{"Recent":[0],"years":[1],"have":[2,17],"seen":[3],"considerable":[4],"advancements":[5],"in":[6,68,161],"multi-step":[7,80],"reasoning":[8,33,85,109,178],"with":[9,43],"Large":[10],"Language":[11],"Models":[12],"(LLMs).":[13],"The":[14,35],"previous":[15],"studies":[16],"elucidated":[18],"the":[19,32,47,64,69,75,101,108,122,137,162,168],"merits":[20],"of":[21,66,124,139,171],"integrating":[22],"feedback":[23,45,103],"or":[24,56],"search":[25,97],"mechanisms":[26],"during":[27,46],"model":[28],"inference":[29,70,176],"to":[30,51,62,72,106,121,135,147,175],"improve":[31],"accuracy.":[34],"Process-Supervised":[36],"Reward":[37],"Model":[38],"(PRM),":[39],"typically":[40],"furnishes":[41],"LLMs":[42],"step-by-step":[44],"training":[48],"phase,":[49],"akin":[50],"Proximal":[52],"Policy":[53],"Optimization":[54],"(PPO)":[55],"reject":[57],"sampling.":[58],"Our":[59],"objective":[60],"is":[61],"examine":[63],"efficacy":[65],"PRM":[67,105,116],"phase":[71],"help":[73],"discern":[74],"optimal":[76],"solution":[77],"paths":[78],"for":[79,153,177],"tasks":[81,155],"such":[82],"as":[83],"mathematical":[84,128],"and":[86,132,156],"code":[87,163],"generation.":[88],"To":[89],"this":[90],"end,":[91],"we":[92,142],"propose":[93],"a":[94,144],"heuristic":[95],"greedy":[96],"algorithm":[98],"that":[99],"employs":[100],"step-level":[102,150],"from":[104],"optimize":[107],"pathways":[110],"explored":[111],"by":[112],"LLMs.":[113],"This":[114],"tailored":[115],"demonstrated":[117],"enhanced":[118],"results":[119],"compared":[120],"Chain":[123],"Thought":[125],"(CoT)":[126],"on":[127],"benchmarks":[129],"like":[130],"GSM8K":[131],"MATH.":[133],"Additionally,":[134],"explore":[136],"versatility":[138],"our":[140,172],"approach,":[141],"develop":[143],"novel":[145],"method":[146],"automatically":[148],"generate":[149],"reward":[151],"dataset":[152],"coding":[154],"observed":[157],"similar":[158],"improved":[159],"performance":[160],"generation":[164],"tasks.":[165,179],"Thus":[166],"highlighting":[167],"robust":[169],"nature":[170],"reward-model-based":[173],"approach":[174]},"counts_by_year":[],"updated_date":"2026-10-14T16:39:01.820549","created_date":"2023-10-18T00:00:00"}