{"id":"https://openalex.org/W4415318142","doi":"https://doi.org/10.48550/arxiv.2510.07147","title":"A Multi-Agent Framework for Stateful Inference-Time Search","display_name":"A Multi-Agent Framework for Stateful Inference-Time Search","publication_year":2025,"publication_date":"2025-10-08","ids":{"openalex":"https://openalex.org/W4415318142","doi":"https://doi.org/10.48550/arxiv.2510.07147"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2510.07147","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.07147","pdf_url":"https://arxiv.org/pdf/2510.07147","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","license_id":"https://openalex.org/licenses/cc-by","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/2510.07147","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5063534334","display_name":"Arshika Lalan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210153414","display_name":"Nutanix (United States)","ror":"https://ror.org/04tz2xx73","country_code":"US","type":"company","lineage":["https://openalex.org/I4210153414"]},{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lalan, Arshika","raw_affiliation_strings":["Carnegie Mellon University","Nutanix"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]},{"raw_affiliation_string":"Nutanix","institution_ids":["https://openalex.org/I4210153414"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043902598","display_name":"Rajat Ghosh","orcid":"https://orcid.org/0000-0003-2919-9693","observed_orcids":["https://orcid.org/0000-0003-2919-9693"]},"institutions":[{"id":"https://openalex.org/I4210153414","display_name":"Nutanix (United States)","ror":"https://ror.org/04tz2xx73","country_code":"US","type":"company","lineage":["https://openalex.org/I4210153414"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ghosh, Rajat","raw_affiliation_strings":["Nutanix"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nutanix","institution_ids":["https://openalex.org/I4210153414"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120050436","display_name":"Aditya Kolsur","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210153414","display_name":"Nutanix (United States)","ror":"https://ror.org/04tz2xx73","country_code":"US","type":"company","lineage":["https://openalex.org/I4210153414"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kolsur, Aditya","raw_affiliation_strings":["Nutanix"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nutanix","institution_ids":["https://openalex.org/I4210153414"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5113845539","display_name":"Debojyoti Dutta","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210153414","display_name":"Nutanix (United States)","ror":"https://ror.org/04tz2xx73","country_code":"US","type":"company","lineage":["https://openalex.org/I4210153414"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dutta, Debojyoti","raw_affiliation_strings":["Nutanix"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nutanix","institution_ids":["https://openalex.org/I4210153414"]}]}],"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":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/T10743","display_name":"Software Testing and Debugging Techniques","score":0.9678000211715698,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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/T10743","display_name":"Software Testing and Debugging Techniques","score":0.9678000211715698,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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/T11975","display_name":"Evolutionary Algorithms and Applications","score":0.02019999921321869,"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/T10260","display_name":"Software Engineering Research","score":0.010400000028312206,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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-agent-systems","display_name":"multi-agent systems","score":0.7020000219345093},{"id":"https://openalex.org/keywords/evolutionary-search","display_name":"evolutionary search","score":0.7480000257492065},{"id":"https://openalex.org/keywords/automated-unit-test-generation","display_name":"automated unit test generation","score":0.8080000281333923},{"id":"https://openalex.org/keywords/persistent-state","display_name":"persistent state","score":0.5580000281333923},{"id":"https://openalex.org/keywords/llm-based-agents","display_name":"LLM agents","score":0.6290000081062317},{"id":"https://openalex.org/keywords/multi-step-reasoning","display_name":"multi-step reasoning","score":0.75},{"id":"https://openalex.org/keywords/test-coverage","display_name":"test coverage","score":0.5019999742507935}],"concepts":[{"id":"https://openalex.org/C103613024","wikidata":"https://www.wikidata.org/wiki/Q230924","display_name":"Stateless protocol","level":3,"score":0.8998000025749207},{"id":"https://openalex.org/C22927095","wikidata":"https://www.wikidata.org/wiki/Q1784206","display_name":"Stateful firewall","level":3,"score":0.8978999853134155},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6863999962806702},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.613099992275238},{"id":"https://openalex.org/C159149176","wikidata":"https://www.wikidata.org/wiki/Q14489129","display_name":"Evolutionary algorithm","level":2,"score":0.4767000079154968},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.41769999265670776},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3950999975204468},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39089998602867126},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.37439998984336853},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.3716999888420105},{"id":"https://openalex.org/C105902424","wikidata":"https://www.wikidata.org/wiki/Q1197129","display_name":"Evolutionary computation","level":2,"score":0.3391999900341034},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.30090001225471497},{"id":"https://openalex.org/C199505168","wikidata":"https://www.wikidata.org/wiki/Q3267529","display_name":"Evolutionary robotics","level":3,"score":0.28439998626708984},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.26080000400543213}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2510.07147","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.07147","pdf_url":"https://arxiv.org/pdf/2510.07147","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","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2510.07147","doi":"https://doi.org/10.48550/arxiv.2510.07147","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.07147","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","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2510.07147","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.07147","pdf_url":"https://arxiv.org/pdf/2510.07147","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","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4415318142.pdf","grobid_xml":"https://content.openalex.org/works/W4415318142.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"work":[1],"explores":[2],"agentic":[3],"inference-time":[4,70,189],"techniques":[5],"to":[6,20],"perform":[7],"structured,":[8],"multi-step":[9,17],"reasoning.":[10],"However,":[11],"stateless":[12,64,159],"inference":[13,151],"often":[14,31],"struggles":[15],"on":[16,39,163],"tasks":[18,40],"due":[19],"the":[21,89],"absence":[22],"of":[23,91,137],"persistent":[24,69,116,188],"state.":[25],"Moreover,":[26],"task-specific":[27],"fine-tuning":[28],"or":[29],"instruction-tuning":[30],"achieve":[32],"surface-level":[33],"code":[34],"generation":[35,87,90],"but":[36],"remain":[37],"brittle":[38],"requiring":[41],"deeper":[42],"reasoning":[43],"and":[44,75,110,125,170,172,181],"long-horizon":[45],"dependencies.":[46],"To":[47],"address":[48],"these":[49],"limitations,":[50],"we":[51],"propose":[52],"stateful":[53,149],"multi-agent":[54,150],"evolutionary":[55,77,101,121,192],"search,":[56],"a":[57,133],"training-free":[58],"framework":[59,152],"that":[60,186],"departs":[61],"from":[62],"prior":[63],"approaches":[65],"by":[66],"combining":[67,187],"(i)":[68],"state,":[71],"(ii)":[72],"adversarial":[73],"mutation,":[74],"(iii)":[76],"preservation.":[78],"We":[79,94],"demonstrate":[80],"its":[81],"effectiveness":[82],"in":[83,156],"automated":[84],"unit":[85],"test":[86],"through":[88],"edge":[92,97,141],"cases.":[93,130],"generate":[95],"robust":[96],"cases":[98,142],"using":[99,173],"an":[100],"search":[102,193],"process,":[103],"where":[104],"specialized":[105],"agents":[106],"sequentially":[107],"propose,":[108],"mutate,":[109],"score":[111],"candidates.":[112],"A":[113],"controller":[114],"maintains":[115],"state":[117,190],"across":[118,127,143],"generations,":[119],"while":[120],"preservation":[122],"ensures":[123],"diversity":[124],"exploration":[126],"all":[128],"possible":[129],"This":[131],"yields":[132],"generalist":[134],"agent":[135],"capable":[136],"discovering":[138],"robust,":[139],"high-coverage":[140],"unseen":[144],"codebases.":[145],"Experiments":[146],"show":[147],"our":[148],"achieves":[153],"substantial":[154],"gains":[155],"coverage":[157],"over":[158],"single-step":[160],"baselines,":[161],"evaluated":[162],"prevalent":[164],"unit-testing":[165],"benchmarks":[166],"such":[167],"as":[168],"HumanEval":[169],"TestGenEvalMini":[171],"three":[174],"diverse":[175],"LLM":[176],"families":[177],"-":[178],"Llama,":[179],"Gemma,":[180],"GPT.":[182],"These":[183],"results":[184],"indicate":[185],"with":[191],"materially":[194],"improves":[195],"unit-test":[196],"generation.":[197]},"counts_by_year":[],"updated_date":"2026-09-30T09:27:27.532063","created_date":"2025-10-18T00:00:00"}