{"id":"https://openalex.org/W7166825679","doi":"https://doi.org/10.18653/v1/2026.findings-acl.736","title":"MemSearcher: Iterative Memory Integration for Search Agent via End-to-End Reinforcement Learning","display_name":"MemSearcher: Iterative Memory Integration for Search Agent via End-to-End Reinforcement Learning","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166825679","doi":"https://doi.org/10.18653/v1/2026.findings-acl.736"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.736","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.736","pdf_url":"https://aclanthology.org/2026.findings-acl.736.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.736.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5021935122","display_name":"Qianhao Yuan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qianhao Yuan","raw_affiliation_strings":["Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China","University of Chinese Academy of Sciences , Beijing , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China","institution_ids":["https://openalex.org/I4210128818"]},{"raw_affiliation_string":"University of Chinese Academy of Sciences , Beijing , China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139822465","display_name":"Jie Lou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jie Lou","raw_affiliation_strings":["Xiaohongshu Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Xiaohongshu Inc","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139758564","display_name":"Zichao Li","orcid":null},"institutions":[{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zichao Li","raw_affiliation_strings":["Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China","University of Chinese Academy of Sciences , Beijing , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China","institution_ids":["https://openalex.org/I4210128818"]},{"raw_affiliation_string":"University of Chinese Academy of Sciences , Beijing , China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139756132","display_name":"Jiawei Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiawei Chen","raw_affiliation_strings":["Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China","University of Chinese Academy of Sciences , Beijing , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China","institution_ids":["https://openalex.org/I4210128818"]},{"raw_affiliation_string":"University of Chinese Academy of Sciences , Beijing , China","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139799701","display_name":"Yaojie Lu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yaojie Lu","raw_affiliation_strings":["Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China","institution_ids":["https://openalex.org/I4210128818"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139813510","display_name":"Hongyu Lin","orcid":null},"institutions":[{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongyu Lin","raw_affiliation_strings":["Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China","institution_ids":["https://openalex.org/I4210128818"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139739506","display_name":"Le Sun","orcid":null},"institutions":[{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Le Sun","raw_affiliation_strings":["Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China","institution_ids":["https://openalex.org/I4210128818"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139779223","display_name":"Debing Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Debing Zhang","raw_affiliation_strings":["Xiaohongshu Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Xiaohongshu Inc","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139718466","display_name":"Xianpei Han","orcid":null},"institutions":[{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xianpei Han","raw_affiliation_strings":["Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Information Processing Laboratory , Institute of Software , Chinese Academy of Sciences , Beijing , China","institution_ids":["https://openalex.org/I4210128818"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.82169081,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"14965","last_page":"14977"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12288","display_name":"Optimization and Search Problems","score":0.07840000092983246,"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"}},"topics":[{"id":"https://openalex.org/T12288","display_name":"Optimization and Search Problems","score":0.07840000092983246,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.05959999933838844,"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/T10742","display_name":"Peer-to-Peer Network Technologies","score":0.03909999877214432,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6425999999046326},{"id":"https://openalex.org/keywords/iterative-learning-control","display_name":"Iterative learning control","score":0.31369999051094055},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.30809998512268066},{"id":"https://openalex.org/keywords/scheme","display_name":"Scheme (mathematics)","score":0.3052000105381012},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.30469998717308044},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.28760001063346863}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6471999883651733},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6425999999046326},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5519999861717224},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3319999873638153},{"id":"https://openalex.org/C117619785","wikidata":"https://www.wikidata.org/wiki/Q6094414","display_name":"Iterative learning control","level":3,"score":0.31369999051094055},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.30809998512268066},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.3052000105381012},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.30469998717308044},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.28760001063346863},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.28540000319480896},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.28439998626708984},{"id":"https://openalex.org/C199190896","wikidata":"https://www.wikidata.org/wiki/Q3509276","display_name":"Learning classifier system","level":3,"score":0.273499995470047},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.25690001249313354},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.736","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.736","pdf_url":"https://aclanthology.org/2026.findings-acl.736.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.736","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.736","pdf_url":"https://aclanthology.org/2026.findings-acl.736.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","score":0.48107320070266724,"display_name":"Reduced inequalities"}],"awards":[{"id":"https://openalex.org/G2823011528","display_name":null,"funder_award_id":"62306303","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166825679.pdf","grobid_xml":"https://content.openalex.org/works/W7166825679.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"LLM-based":[0],"search":[1],"agents":[2],"often":[3],"concatenate":[4],"the":[5,10,49],"full":[6],"interaction":[7],"history":[8],"into":[9],"context,":[11],"producing":[12],"long":[13],"and":[14,17,21,46,115],"noisy":[15],"inputs,":[16],"increasing":[18],"compute":[19],"cost":[20],"GPU":[22],"memory":[23,38],"overhead.To":[24],"address":[25],"this":[26],"issue,":[27],"we":[28],"propose":[29],"Mem-Searcher,":[30],"an":[31,71],"agent":[32],"framework":[33],"that":[34,92],"maintains":[35],"a":[36,100],"compact":[37],"during":[39],"multi-turn":[40],"interactions,":[41],"retaining":[42],"only":[43],"question-relevant":[44],"information":[45],"thereby":[47],"keeping":[48],"context":[50],"length":[51],"stable":[52],"across":[53,111],"turns.Training":[54],"MemSearcher":[55,93],"is":[56],"challenging":[57],"because":[58],"each":[59,69],"trajectory":[60],"spans":[61],"multiple":[62],"turns":[63,87],"under":[64],"different":[65],"LLM":[66],"contexts,":[67],"making":[68],"turn":[70],"independent":[72],"optimization":[73],"target":[74],"in":[75],"reinforcement":[76],"learning.We":[77],"introduce":[78],"multi-context":[79],"GRPO,":[80],"which":[81],"propagates":[82],"trajectory-level":[83],"advantages":[84],"to":[85],"all":[86],"for":[88],"end-to-end":[89],"optimization.Experiments":[90],"demonstrate":[91],"outperforms":[94],"strong":[95],"history-concatenation":[96],"(ReAct-style)":[97],"baselines":[98],"on":[99],"range":[101],"of":[102],"public":[103],"datasets":[104],"while":[105],"maintaining":[106],"nearly":[107],"constant":[108],"token":[109],"counts":[110],"multiturn":[112],"interactions.The":[113],"code":[114],"models":[116],"will":[117],"be":[118],"publicly":[119],"available":[120],"at":[121],"https://github.com/icip-":[122],"cas/MemSearcher.":[123]},"counts_by_year":[],"updated_date":"2026-08-24T07:32:12.397491","created_date":"2026-07-02T00:00:00"}