{"id":"https://openalex.org/W7140181296","doi":"https://doi.org/10.48550/arxiv.2603.20212","title":"Fast-Slow Thinking RM: Efficient Integration of Scalar and Generative Reward Models","display_name":"Fast-Slow Thinking RM: Efficient Integration of Scalar and Generative Reward Models","publication_year":2026,"publication_date":"2026-03-02","ids":{"openalex":"https://openalex.org/W7140181296","doi":"https://doi.org/10.48550/arxiv.2603.20212"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.20212","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.20212","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.20212","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Wu, Jiayun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Jiayun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Hou, Peixu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hou, Peixu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Qu, Shan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qu, Shan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Zhang, Peng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Peng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Gu, Ning","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gu, Ning","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Lu, Tun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Tun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.19030000269412994,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.19030000269412994,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.09839999675750732,"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/T10028","display_name":"Topic Modeling","score":0.06589999794960022,"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/adaptability","display_name":"Adaptability","score":0.6226999759674072},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.5655999779701233},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5619999766349792},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.4147999882698059},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.4101000130176544},{"id":"https://openalex.org/keywords/scalar","display_name":"Scalar (mathematics)","score":0.39980000257492065},{"id":"https://openalex.org/keywords/dual","display_name":"Dual (grammatical number)","score":0.38350000977516174},{"id":"https://openalex.org/keywords/temporal-difference-learning","display_name":"Temporal difference learning","score":0.3776000142097473}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6583999991416931},{"id":"https://openalex.org/C177606310","wikidata":"https://www.wikidata.org/wiki/Q5674297","display_name":"Adaptability","level":2,"score":0.6226999759674072},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.5655999779701233},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5619999766349792},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.48910000920295715},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4341000020503998},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.4147999882698059},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4101000130176544},{"id":"https://openalex.org/C57691317","wikidata":"https://www.wikidata.org/wiki/Q1289248","display_name":"Scalar (mathematics)","level":2,"score":0.39980000257492065},{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.38350000977516174},{"id":"https://openalex.org/C196340769","wikidata":"https://www.wikidata.org/wiki/Q7698910","display_name":"Temporal difference learning","level":3,"score":0.3776000142097473},{"id":"https://openalex.org/C190839683","wikidata":"https://www.wikidata.org/wiki/Q2448197","display_name":"Train","level":2,"score":0.36329999566078186},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.3529999852180481},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.34139999747276306},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.33340001106262207},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.3296000063419342},{"id":"https://openalex.org/C2779231336","wikidata":"https://www.wikidata.org/wiki/Q7534724","display_name":"Sketch","level":2,"score":0.30880001187324524},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.2903999984264374},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2777000069618225},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.25270000100135803}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.20212","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.20212","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.20212","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.20212","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reward":[0,19,36,55],"models":[1,113],"(RMs)":[2],"are":[3],"critical":[4],"for":[5],"aligning":[6],"Large":[7],"Language":[8],"Models":[9,20,37,56],"via":[10],"Reinforcement":[11],"Learning":[12],"from":[13,43],"Human":[14],"Feedback":[15],"(RLHF).":[16],"While":[17],"Generative":[18],"(GRMs)":[21],"achieve":[22],"superior":[23],"accuracy":[24],"through":[25],"chain-of-thought":[26],"(CoT)":[27],"reasoning,":[28],"they":[29],"incur":[30],"substantial":[31],"computational":[32],"costs.":[33],"Conversely,":[34],"Scalar":[35],"(SRMs)":[38],"offer":[39],"efficiency":[40],"but":[41],"suffer":[42],"limited":[44],"performance":[45,109],"and":[46,86,121],"adaptability":[47],"in":[48],"complex":[49],"scenarios.":[50],"We":[51],"introduce":[52],"Fast-Slow":[53],"Thinking":[54],"(F/S-RM),":[57],"a":[58,69,81,93,106],"hybrid":[59],"RM":[60],"architecture":[61],"inspired":[62],"by":[63,92,118],"Dual":[64],"Process":[65],"Theory.":[66],"It":[67],"trains":[68],"single":[70],"model":[71],"to":[72,100],"integrate":[73],"two":[74],"distinct":[75],"reward":[76],"paradigms:":[77],"first-token":[78],"prediction":[79],"as":[80],"scalar":[82],"score":[83],"(fast":[84],"thinking)":[85],"CoT-based":[87],"judgment":[88],"(slow":[89],"thinking),":[90],"regulated":[91],"dual-confidence":[94],"activation":[95],"mechanism":[96],"that":[97],"determines":[98],"when":[99],"activate":[101],"slow":[102],"thinking.":[103],"F/S-RM":[104],"achieves":[105],"1.2%":[107],"relative":[108],"improvement":[110],"over":[111],"state-of-the-art":[112],"while":[114],"reducing":[115],"token":[116],"consumption":[117],"20.8%.":[119],"Code":[120],"data":[122],"will":[123],"be":[124],"publicly":[125],"available.":[126]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-03-25T00:00:00"}