{"id":"https://openalex.org/W7160407713","doi":"https://doi.org/10.48550/arxiv.2605.02933","title":"Relation Reasoning with LLMs in Expensive Optimization","display_name":"Relation Reasoning with LLMs in Expensive Optimization","publication_year":2026,"publication_date":"2026-04-30","ids":{"openalex":"https://openalex.org/W7160407713","doi":"https://doi.org/10.48550/arxiv.2605.02933"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.02933","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.02933","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.2605.02933","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101138518","display_name":"Ye Lu","orcid":"https://orcid.org/0009-0009-4948-9680"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Ye","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135482248","display_name":"Bingdong Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Bingdong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135485017","display_name":"Aimin Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Aimin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135445125","display_name":"Hao Hao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao, Hao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"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/T10848","display_name":"Advanced Multi-Objective Optimization Algorithms","score":0.7254999876022339,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T10848","display_name":"Advanced Multi-Objective Optimization Algorithms","score":0.7254999876022339,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T12535","display_name":"Machine Learning and Data Classification","score":0.09629999846220016,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.023800000548362732,"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/inference","display_name":"Inference","score":0.5273000001907349},{"id":"https://openalex.org/keywords/pairwise-comparison","display_name":"Pairwise comparison","score":0.5087000131607056},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.4447000026702881},{"id":"https://openalex.org/keywords/evolutionary-algorithm","display_name":"Evolutionary algorithm","score":0.4318999946117401},{"id":"https://openalex.org/keywords/evolutionary-computation","display_name":"Evolutionary computation","score":0.4124000072479248},{"id":"https://openalex.org/keywords/optimization-problem","display_name":"Optimization problem","score":0.3995000123977661},{"id":"https://openalex.org/keywords/population","display_name":"Population","score":0.3813999891281128},{"id":"https://openalex.org/keywords/relation","display_name":"Relation (database)","score":0.3603000044822693}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6021999716758728},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5273000001907349},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.513700008392334},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.5087000131607056},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4708000123500824},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46540001034736633},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.4447000026702881},{"id":"https://openalex.org/C159149176","wikidata":"https://www.wikidata.org/wiki/Q14489129","display_name":"Evolutionary algorithm","level":2,"score":0.4318999946117401},{"id":"https://openalex.org/C105902424","wikidata":"https://www.wikidata.org/wiki/Q1197129","display_name":"Evolutionary computation","level":2,"score":0.4124000072479248},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.3995000123977661},{"id":"https://openalex.org/C2908647359","wikidata":"https://www.wikidata.org/wiki/Q2625603","display_name":"Population","level":2,"score":0.3813999891281128},{"id":"https://openalex.org/C25343380","wikidata":"https://www.wikidata.org/wiki/Q277521","display_name":"Relation (database)","level":2,"score":0.3603000044822693},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.34459999203681946},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.32249999046325684},{"id":"https://openalex.org/C129844170","wikidata":"https://www.wikidata.org/wiki/Q41299","display_name":"Quadratic equation","level":2,"score":0.2937000095844269},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.29260000586509705},{"id":"https://openalex.org/C131675550","wikidata":"https://www.wikidata.org/wiki/Q7646884","display_name":"Surrogate model","level":2,"score":0.2703000009059906},{"id":"https://openalex.org/C55660270","wikidata":"https://www.wikidata.org/wiki/Q5164377","display_name":"Constrained optimization","level":2,"score":0.260699987411499},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.25200000405311584},{"id":"https://openalex.org/C112313634","wikidata":"https://www.wikidata.org/wiki/Q7886648","display_name":"Complement (music)","level":5,"score":0.25}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.02933","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.02933","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.2605.02933","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.02933","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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Expensive":[0],"optimization":[1,135],"problems":[2],"(EOPs)":[3],"are":[4,161],"black-box":[5],"tasks":[6],"with":[7,121],"costly":[8],"objective":[9],"evaluations":[10,27],"and":[11,95,118,126,133,141,159],"no":[12],"gradient":[13],"access,":[14],"making":[15],"the":[16,19],"evaluation":[17],"budget":[18],"key":[20],"bottleneck.":[21],"Surrogate-assisted":[22],"evolutionary":[23,56,74,116],"algorithms":[24],"(SAEAs)":[25],"reduce":[26],"via":[28],"surrogate":[29,54,61,153],"predictions,":[30],"but":[31],"conventional":[32],"surrogates":[33],"often":[34],"require":[35],"frequent":[36],"retraining":[37],"as":[38,63],"populations":[39],"evolve,":[40],"incurring":[41],"overhead.":[42],"This":[43],"paper":[44],"proposes":[45],"R2SAEA,":[46],"a":[47,96,151],"reinforcement-trained":[48],"relation-based":[49,60],"large":[50],"language":[51],"model":[52],"(LLM)":[53],"assisted":[55],"algorithm.":[57],"We":[58,109],"cast":[59],"modeling":[62],"an":[64,78,112],"in-context":[65],"pairwise":[66],"reasoning":[67],"task.":[68],"To":[69],"enable":[70],"efficient":[71,147],"inference":[72],"in":[73,92],"loops,":[75],"we":[76],"develop":[77],"anchor-based":[79],"iterative":[80],"context":[81],"construction":[82],"strategy":[83],"that":[84,100],"reduces":[85],"prompt":[86],"complexity":[87],"from":[88,115],"quadratic":[89],"to":[90],"linear":[91],"population":[93],"size,":[94],"voting-based":[97],"aggregation":[98],"scheme":[99],"converts":[101],"predicted":[102],"relations":[103],"into":[104],"scores":[105],"for":[106],"offspring":[107],"selection.":[108],"further":[110],"build":[111],"RL":[113],"pipeline":[114],"trajectories":[117],"fine-tune":[119],"Qwen2.5":[120],"GRPO.":[122],"Experiments":[123],"on":[124],"single-":[125],"multi-objective":[127],"benchmarks":[128],"show":[129],"improved":[130],"relation":[131],"prediction":[132],"state-of-the-art":[134],"performance":[136],"over":[137],"strong":[138],"SAEA":[139],"baselines":[140],"general":[142],"LLMs.":[143],"Quantization":[144],"also":[145],"enables":[146],"edge":[148],"deployment,":[149],"supporting":[150],"zero-shot":[152],"paradigm":[154],"without":[155],"per-generation":[156],"retraining.":[157],"Code":[158],"models":[160],"available":[162],"at":[163],"https://github.com/Septend9/R2SAEA.":[164]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-07T00:00:00"}