{"id":"https://openalex.org/W7212129377","doi":"https://doi.org/10.48550/arxiv.2609.09957","title":"Beyond Verified Answers: Solver-Informed Self-Distillation for Bootstrapping Operations Research Language Models","display_name":"Beyond Verified Answers: Solver-Informed Self-Distillation for Bootstrapping Operations Research Language Models","publication_year":2026,"publication_date":"2026-09-09","ids":{"openalex":"https://openalex.org/W7212129377","doi":"https://doi.org/10.48550/arxiv.2609.09957"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.09957","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.09957","pdf_url":"https://arxiv.org/pdf/2609.09957","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.09957","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5151204782","display_name":"Rui Zhu","orcid":null,"observed_orcids":[]},"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":"Zhu, Rui","raw_affiliation_strings":["Shanghai Jiao Tong University , Shanghai , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University , Shanghai , China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081931664","display_name":"\u66f9\u660e\u9f99","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cao, Minglong","raw_affiliation_strings":["Fudan University , Shanghai , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fudan University , Shanghai , China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5151265384","display_name":"Chenyu Zhou","orcid":null,"observed_orcids":[]},"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":"Zhou, Chenyu","raw_affiliation_strings":["Shanghai Jiao Tong University , Shanghai , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University , Shanghai , China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140291035","display_name":"Jianghao Lin","orcid":null,"observed_orcids":[]},"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":"Lin, Jianghao","raw_affiliation_strings":["Shanghai Jiao Tong University , Shanghai , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University , Shanghai , China","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5140238768","display_name":"Dongdong Ge","orcid":null,"observed_orcids":[]},"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":"Ge, Dongdong","raw_affiliation_strings":["Shanghai Jiao Tong University , Shanghai , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University , Shanghai , China","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"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":0.5785999894142151,"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":0.5785999894142151,"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":0.09560000151395798,"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/T13904","display_name":"Artificial Intelligence Applications","score":0.04569999873638153,"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/self-distillation","display_name":"self-distillation","score":0.8980000019073486},{"id":"https://openalex.org/keywords/on-policy-learning","display_name":"on-policy learning","score":0.5910000205039978},{"id":"https://openalex.org/keywords/credit-assignment","display_name":"credit assignment","score":0.6970000267028809}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8116000294685364},{"id":"https://openalex.org/C207609745","wikidata":"https://www.wikidata.org/wiki/Q4944086","display_name":"Bootstrapping (finance)","level":2,"score":0.7228999733924866},{"id":"https://openalex.org/C2779010991","wikidata":"https://www.wikidata.org/wiki/Q2720909","display_name":"Artifact (error)","level":2,"score":0.7152000069618225},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6869999766349792},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6118999719619751},{"id":"https://openalex.org/C2778770139","wikidata":"https://www.wikidata.org/wiki/Q1966904","display_name":"Solver","level":2,"score":0.578000009059906},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5730000138282776},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5429999828338623},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5174999833106995},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4596000015735626},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.4575999975204468},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.3686999976634979},{"id":"https://openalex.org/C164155591","wikidata":"https://www.wikidata.org/wiki/Q2067766","display_name":"Satisfiability modulo theories","level":2,"score":0.3452000021934509},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.32749998569488525},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.325300008058548},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.30570000410079956},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2824999988079071},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.27720001339912415},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2689000070095062},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.2637999951839447}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.09957","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.09957","pdf_url":"https://arxiv.org/pdf/2609.09957","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.09957","doi":"https://doi.org/10.48550/arxiv.2609.09957","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.09957","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-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.09957","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.09957","pdf_url":"https://arxiv.org/pdf/2609.09957","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7212129377.pdf","grobid_xml":"https://content.openalex.org/works/W7212129377.grobid-xml"},"referenced_works_count":19,"referenced_works":["https://openalex.org/W2736601468","https://openalex.org/W3177813494","https://openalex.org/W4391631327","https://openalex.org/W4401201897","https://openalex.org/W4403883066","https://openalex.org/W4410192806","https://openalex.org/W4411113222","https://openalex.org/W4412886918","https://openalex.org/W4414939355","https://openalex.org/W7125861034","https://openalex.org/W7126034091","https://openalex.org/W7140517510","https://openalex.org/W7153056532","https://openalex.org/W7154507714","https://openalex.org/W7157650971","https://openalex.org/W7162554671","https://openalex.org/W7162663127","https://openalex.org/W7164364162","https://openalex.org/W7167642481"],"related_works":["https://openalex.org/W7083692996","https://openalex.org/W7155098142","https://openalex.org/W4414814642","https://openalex.org/W7134925491","https://openalex.org/W7214042457","https://openalex.org/W7172387483","https://openalex.org/W4416235990","https://openalex.org/W7166525465","https://openalex.org/W7127107409","https://openalex.org/W7164740129"],"abstract_inverted_index":{"Modern":[0],"large":[1],"language":[2,134],"models":[3,135,185],"(LLMs)":[4],"can":[5,84,100,196],"translate":[6],"natural-language":[7],"descriptions":[8],"into":[9],"operations":[10],"research":[11],"(OR)":[12],"formulations.":[13,36],"Post-training":[14],"techniques":[15],"including":[16],"reinforcement":[17],"learning":[18],"and":[19,152,168,183],"on-policy":[20],"self-distillation":[21,83,111],"have":[22],"further":[23],"improved":[24],"this":[25],"capability.":[26],"However,":[27],"three":[28],"limitations":[29],"remain":[30],"in":[31],"training":[32,38],"LLMs":[33],"for":[34,131,180],"OR":[35,133,174],"First,":[37],"commonly":[39],"relies":[40],"on":[41],"synthetic":[42],"formulations":[43],"validated":[44],"by":[45,88,106],"human":[46],"experts":[47],"or":[48,60,139],"stronger":[49],"models,":[50],"constraining":[51],"scalable":[52,198],"supervision.":[53,118],"Second,":[54],"credit":[55],"assignment":[56],"is":[57],"either":[58],"coarse":[59],"costly:":[61],"outcome":[62],"rewards":[63],"score":[64],"an":[65,78],"entire":[66],"trajectory":[67],"without":[68,136,200],"locating":[69],"the":[70],"responsible":[71],"modeling":[72],"decision,":[73],"whereas":[74],"process-level":[75],"supervision":[76],"requires":[77],"additional":[79],"evaluator.":[80],"Third,":[81],"privileged":[82],"induce":[85],"style":[86],"mismatch":[87],"using":[89,165],"solver":[90,194],"context":[91],"unavailable":[92],"at":[93],"deployment.":[94],"We":[95],"find":[96],"that":[97,193],"a":[98,112,128,154,158],"model":[99,161],"improve":[101],"from":[102,146],"solver-artifact":[103],"feedback":[104],"generated":[105],"its":[107],"own":[108],"rollouts,":[109,148],"making":[110],"practical,":[113],"evaluator-free":[114],"source":[115],"of":[116],"dense":[117,169],"Therefore,":[119],"we":[120],"propose":[121],"SOLID:":[122],"Solver-Informed":[123],"On-Policy":[124],"LearnIng":[125],"through":[126],"Self-Distillation,":[127],"novel":[129],"framework":[130],"self-improving":[132],"verified":[137],"answers":[138],"external":[140],"evaluators.":[141],"SOLID":[142,176],"executes":[143],"candidate":[144],"programs":[145],"multiple":[147,173],"clusters":[149],"their":[150],"objectives,":[151],"selects":[153],"majority-group":[155],"artifact":[156],"as":[157],"pseudo-reference.":[159],"The":[160],"then":[162],"performs":[163],"updates":[164],"group-relative":[166,188],"advantages":[167],"self-supervision":[170],"signals.":[171],"Across":[172],"benchmarks,":[175],"improves":[177],"solution":[178],"accuracy":[179],"both":[181],"general-purpose":[182],"OR-tuned":[184],"over":[186],"outcome-only":[187],"training.":[189],"These":[190],"results":[191],"show":[192],"artifacts":[195],"support":[197],"self-improvement":[199],"trusted":[201],"answers.":[202]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-11T00:00:00"}