{"id":"https://openalex.org/W7147506281","doi":"https://doi.org/10.48550/arxiv.2603.29493","title":"MemFactory: Unified Inference & Training Framework for Agent Memory","display_name":"MemFactory: Unified Inference & Training Framework for Agent Memory","publication_year":2026,"publication_date":"2026-03-31","ids":{"openalex":"https://openalex.org/W7147506281","doi":"https://doi.org/10.48550/arxiv.2603.29493"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.29493","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.29493","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.29493","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5132681354","display_name":"Ziliang Guo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Ziliang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102983694","display_name":"Ziheng Li","orcid":"https://orcid.org/0000-0002-3678-261X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Ziheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Tang, Bo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tang, Bo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Xiong, Feiyu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiong, Feiyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Li, Zhiyu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Zhiyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.4887999892234802,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.4887999892234802,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.0737999975681305,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.0551999993622303,"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/modular-design","display_name":"Modular design","score":0.636900007724762},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6241999864578247},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6051999926567078},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.5530999898910522},{"id":"https://openalex.org/keywords/implementation","display_name":"Implementation","score":0.5238000154495239},{"id":"https://openalex.org/keywords/memory-model","display_name":"Memory model","score":0.3806999921798706},{"id":"https://openalex.org/keywords/modularity","display_name":"Modularity (biology)","score":0.353300005197525},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.3521000146865845}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8084999918937683},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.636900007724762},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6241999864578247},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6051999926567078},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.5530999898910522},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.5238000154495239},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5006999969482422},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4083999991416931},{"id":"https://openalex.org/C12186640","wikidata":"https://www.wikidata.org/wiki/Q6815743","display_name":"Memory model","level":3,"score":0.3806999921798706},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.376800000667572},{"id":"https://openalex.org/C2779478453","wikidata":"https://www.wikidata.org/wiki/Q6889748","display_name":"Modularity (biology)","level":2,"score":0.353300005197525},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.3521000146865845},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.3375000059604645},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.29409998655319214},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.28290000557899475},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.2815999984741211},{"id":"https://openalex.org/C46743427","wikidata":"https://www.wikidata.org/wiki/Q1341685","display_name":"Inference engine","level":3,"score":0.27869999408721924},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.27469998598098755},{"id":"https://openalex.org/C158600405","wikidata":"https://www.wikidata.org/wiki/Q5054566","display_name":"Causal inference","level":2,"score":0.27000001072883606},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.257099986076355}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.29493","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.29493","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.29493","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.29493","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"score":0.661973237991333,"display_name":"Industry, innovation and infrastructure","id":"https://metadata.un.org/sdg/9"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Memory-augmented":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs)":[4],"are":[5],"essential":[6],"for":[7,77,136,201],"developing":[8],"capable,":[9],"long-term":[10],"AI":[11,206],"agents.":[12,79,207],"Recently,":[13],"applying":[14],"Reinforcement":[15],"Learning":[16],"(RL)":[17],"to":[18,48,101,121,181,196],"optimize":[19],"memory":[20,93,105,124],"operations,":[21],"such":[22],"as":[23,30],"extraction,":[24],"updating,":[25],"and":[26,42,53,72,143,159,188],"retrieval,":[27],"has":[28],"emerged":[29],"a":[31,45,108,185],"highly":[32,40,69],"promising":[33],"research":[34],"direction.":[35],"However,":[36],"existing":[37],"implementations":[38],"remain":[39],"fragmented":[41],"task-specific,":[43],"lacking":[44],"unified":[46,85],"infrastructure":[47],"streamline":[49],"the":[50,66,82,92,112,150,163,170,194,199],"integration,":[51],"training,":[52],"evaluation":[54,160,164],"of":[55,84,179],"these":[56],"complex":[57],"pipelines.":[58],"To":[59],"address":[60],"this":[61],"gap,":[62],"we":[63],"present":[64],"MemFactory,":[65],"first":[67],"unified,":[68],"modular":[70],"training":[71,158],"inference":[73],"framework":[74,113],"specifically":[75],"designed":[76],"memory-augmented":[78],"Inspired":[80],"by":[81,128],"success":[83],"fine-tuning":[86],"frameworks":[87],"like":[88],"LLaMA-Factory,":[89],"MemFactory":[90,132,148,166,191],"abstracts":[91],"lifecycle":[94],"into":[95],"atomic,":[96],"plug-and-play":[97],"components,":[98],"enabling":[99],"researchers":[100],"seamlessly":[102],"construct":[103],"custom":[104],"agents":[106],"via":[107],"\"Lego-like\"":[109],"architecture.":[110],"Furthermore,":[111],"natively":[114],"integrates":[115],"Group":[116],"Relative":[117],"Policy":[118],"Optimization":[119],"(GRPO)":[120],"fine-tune":[122],"internal":[123],"management":[125],"policies":[126],"driven":[127],"multi-dimensional":[129],"environmental":[130],"rewards.":[131],"provides":[133],"out-of-the-box":[134],"support":[135],"recent":[137],"cutting-edge":[138],"paradigms,":[139],"including":[140],"Memory-R1,":[141],"RMM,":[142],"MemAgent.":[144],"We":[145],"empirically":[146],"validate":[147],"on":[149,174],"open-source":[151],"MemAgent":[152],"architecture":[153],"using":[154],"its":[155],"publicly":[156],"available":[157],"data.":[161],"Across":[162],"sets,":[165],"improves":[167],"performance":[168],"over":[169],"corresponding":[171],"base":[172],"models":[173],"average,":[175],"with":[176],"relative":[177],"gains":[178],"up":[180],"14.8%.":[182],"By":[183],"providing":[184],"standardized,":[186],"extensible,":[187],"easy-to-use":[189],"infrastructure,":[190],"significantly":[192],"lowers":[193],"barrier":[195],"entry,":[197],"paving":[198],"way":[200],"future":[202],"innovations":[203],"in":[204],"memory-driven":[205]},"counts_by_year":[],"updated_date":"2026-07-21T08:15:58.654021","created_date":"2026-04-02T00:00:00"}