{"id":"https://openalex.org/W4416962804","doi":"https://doi.org/10.1109/icumt67815.2025.11268643","title":"Reinforcement Learning for Mathematical Reasoning in Small-Scale Language Models with Structured Policy Optimization","display_name":"Reinforcement Learning for Mathematical Reasoning in Small-Scale Language Models with Structured Policy Optimization","publication_year":2025,"publication_date":"2025-11-03","ids":{"openalex":"https://openalex.org/W4416962804","doi":"https://doi.org/10.1109/icumt67815.2025.11268643"},"language":null,"primary_location":{"id":"doi:10.1109/icumt67815.2025.11268643","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icumt67815.2025.11268643","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 17th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101321576","display_name":"Naman Tyagi","orcid":null},"institutions":[{"id":"https://openalex.org/I191972202","display_name":"Amity University","ror":"https://ror.org/02n9z0v62","country_code":"IN","type":"education","lineage":["https://openalex.org/I191972202"]},{"id":"https://openalex.org/I4210126505","display_name":"Amity University","ror":"https://ror.org/02exxtn84","country_code":"AE","type":"education","lineage":["https://openalex.org/I191972202","https://openalex.org/I4210126505"]}],"countries":["AE","IN"],"is_corresponding":true,"raw_author_name":"Naman Tyagi","raw_affiliation_strings":["Amity University Uttar Pradesh,Amity School of Engineering & Technology,Dept. of CSE,Noida,India"],"affiliations":[{"raw_affiliation_string":"Amity University Uttar Pradesh,Amity School of Engineering & Technology,Dept. of CSE,Noida,India","institution_ids":["https://openalex.org/I191972202","https://openalex.org/I4210126505"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005485404","display_name":"Rakesh Chandra Joshi","orcid":"https://orcid.org/0000-0003-1264-9010"},"institutions":[{"id":"https://openalex.org/I191972202","display_name":"Amity University","ror":"https://ror.org/02n9z0v62","country_code":"IN","type":"education","lineage":["https://openalex.org/I191972202"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Rakesh Chandra Joshi","raw_affiliation_strings":["Amity University Uttar Pradesh,Amity Centre for Artificial Intelligence,Noida,India"],"affiliations":[{"raw_affiliation_string":"Amity University Uttar Pradesh,Amity Centre for Artificial Intelligence,Noida,India","institution_ids":["https://openalex.org/I191972202"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Srishti Das","orcid":null},"institutions":[{"id":"https://openalex.org/I191972202","display_name":"Amity University","ror":"https://ror.org/02n9z0v62","country_code":"IN","type":"education","lineage":["https://openalex.org/I191972202"]},{"id":"https://openalex.org/I4210126505","display_name":"Amity University","ror":"https://ror.org/02exxtn84","country_code":"AE","type":"education","lineage":["https://openalex.org/I191972202","https://openalex.org/I4210126505"]}],"countries":["AE","IN"],"is_corresponding":false,"raw_author_name":"Srishti Das","raw_affiliation_strings":["Amity University Uttar Pradesh,Amity School of Engineering & Technology,Dept. of CSE,Noida,India"],"affiliations":[{"raw_affiliation_string":"Amity University Uttar Pradesh,Amity School of Engineering & Technology,Dept. of CSE,Noida,India","institution_ids":["https://openalex.org/I191972202","https://openalex.org/I4210126505"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049703565","display_name":"Pavel Sikora","orcid":"https://orcid.org/0000-0001-5000-4470"},"institutions":[{"id":"https://openalex.org/I60587646","display_name":"Brno University of Technology","ror":"https://ror.org/03613d656","country_code":"CZ","type":"education","lineage":["https://openalex.org/I60587646"]}],"countries":["CZ"],"is_corresponding":false,"raw_author_name":"Pavel Sikora","raw_affiliation_strings":["FEEC, Brno University of Technology,Dept. of Telecommunications,Brno,Czech Republic"],"affiliations":[{"raw_affiliation_string":"FEEC, Brno University of Technology,Dept. of Telecommunications,Brno,Czech Republic","institution_ids":["https://openalex.org/I60587646"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030796738","display_name":"Vojt\u011bch My\u0161ka","orcid":"https://orcid.org/0000-0002-9364-2330"},"institutions":[{"id":"https://openalex.org/I60587646","display_name":"Brno University of Technology","ror":"https://ror.org/03613d656","country_code":"CZ","type":"education","lineage":["https://openalex.org/I60587646"]}],"countries":["CZ"],"is_corresponding":false,"raw_author_name":"Vojtech Myska","raw_affiliation_strings":["FEEC, Brno University of Technology,Dept. of Telecommunications,Brno,Czech Republic"],"affiliations":[{"raw_affiliation_string":"FEEC, Brno University of Technology,Dept. of Telecommunications,Brno,Czech Republic","institution_ids":["https://openalex.org/I60587646"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5038343229","display_name":"Malay Kishore Dutta","orcid":"https://orcid.org/0000-0003-2462-737X"},"institutions":[{"id":"https://openalex.org/I191972202","display_name":"Amity University","ror":"https://ror.org/02n9z0v62","country_code":"IN","type":"education","lineage":["https://openalex.org/I191972202"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Malay Kishore Dutta","raw_affiliation_strings":["Amity University Uttar Pradesh,Amity Centre for Artificial Intelligence,Noida,India"],"affiliations":[{"raw_affiliation_string":"Amity University Uttar Pradesh,Amity Centre for Artificial Intelligence,Noida,India","institution_ids":["https://openalex.org/I191972202"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5101321576"],"corresponding_institution_ids":["https://openalex.org/I191972202","https://openalex.org/I4210126505"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.37682479,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"272","last_page":"277"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.14730000495910645,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.14730000495910645,"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"}},{"id":"https://openalex.org/T11596","display_name":"Constraint Satisfaction and Optimization","score":0.10849999636411667,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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":0.1023000031709671,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.7741000056266785},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.58160001039505},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5511000156402588},{"id":"https://openalex.org/keywords/dual","display_name":"Dual (grammatical number)","score":0.4627000093460083},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.44859999418258667},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.37389999628067017}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7741000056266785},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7580000162124634},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6057999730110168},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.58160001039505},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5511000156402588},{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.4627000093460083},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.46140000224113464},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.44859999418258667},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.37389999628067017},{"id":"https://openalex.org/C76969082","wikidata":"https://www.wikidata.org/wiki/Q486902","display_name":"Mathematical model","level":2,"score":0.37310001254081726},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.3407000005245209},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.3212999999523163},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.28380000591278076},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.25029999017715454}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icumt67815.2025.11268643","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icumt67815.2025.11268643","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 17th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":3,"referenced_works":["https://openalex.org/W3188220908","https://openalex.org/W4387321091","https://openalex.org/W4393763811"],"related_works":[],"abstract_inverted_index":{"Advancing":[0],"mathematical":[1,82,145],"reasoning":[2,64,116,146],"in":[3,144,159],"small-scale":[4,132],"language":[5],"models":[6,50,158],"remains":[7],"a":[8,33,71,109,114,125,151],"challenge":[9],"due":[10],"to":[11,61,75],"their":[12],"limited":[13],"capacity":[14],"and":[15,69,81,97,103],"the":[16],"high":[17],"computational":[18],"demands":[19],"of":[20,48],"standard":[21],"reinforcement":[22,138],"learning":[23],"methods":[24],"like":[25],"Proximal":[26],"Policy":[27,41],"Optimization":[28,42],"(PPO).":[29],"To":[30],"address":[31],"this,":[32],"resource-efficient":[34],"training":[35,85,122],"pipeline":[36],"based":[37],"on":[38,113,124],"Group":[39],"Relative":[40],"(GRPO)":[43],"is":[44],"proposed,":[45],"enabling":[46],"fine-tuning":[47],"compact":[49],"under":[51],"strict":[52],"memory":[53],"constraints.":[54],"The":[55,84,148],"proposed":[56],"method":[57],"introduces":[58],"structured":[59,137],"prompting":[60],"explicitly":[62],"separate":[63],"steps":[65],"from":[66],"final":[67],"answers":[68],"applies":[70],"dual":[72],"reward":[73],"system":[74],"jointly":[76],"optimize":[77],"for":[78,100,154],"format":[79],"adherence":[80],"correctness.":[83],"incorporates":[86],"low-overhead":[87],"techniques":[88],"such":[89],"as":[90],"8-bit":[91],"optimization,":[92],"mixed-precision":[93],"training,":[94],"gradient":[95],"checkpointing,":[96],"accelerated":[98],"decoding":[99],"efficient":[101],"rollout":[102],"policy":[104],"updates.":[105],"Experimental":[106],"results":[107],"show":[108],"$50.95":[110],"\\%$":[111],"accuracy":[112],"benchmark":[115],"dataset\u2014GSM8K,":[117],"outperforming":[118],"several":[119],"larger":[120],"models\u2014while":[121],"entirely":[123],"single":[126],"GPU.":[127],"These":[128],"findings":[129],"demonstrate":[130],"that":[131],"models,":[133],"when":[134],"trained":[135],"with":[136],"learning,":[139],"can":[140],"achieve":[141],"competitive":[142],"performance":[143],"tasks.":[147],"approach":[149],"offers":[150],"practical":[152],"pathway":[153],"deploying":[155],"interpretable,":[156],"reasoning-capable":[157],"low-resource":[160],"environments.":[161]},"counts_by_year":[],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2025-12-03T00:00:00"}