{"id":"https://openalex.org/W7166849228","doi":"https://doi.org/10.18653/v1/2026.findings-acl.648","title":"Reflective RAG: Self-Evaluation Driven Strategy Optimization in Agentic Retrieval-Augmented Generation","display_name":"Reflective RAG: Self-Evaluation Driven Strategy Optimization in Agentic Retrieval-Augmented Generation","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166849228","doi":"https://doi.org/10.18653/v1/2026.findings-acl.648"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.648","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.648","pdf_url":"https://aclanthology.org/2026.findings-acl.648.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.648.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5019019685","display_name":"Haiyan Wu","orcid":"https://orcid.org/0000-0001-8869-6636"},"institutions":[{"id":"https://openalex.org/I90727586","display_name":"Zhejiang University of Finance and Economics","ror":"https://ror.org/055vj5234","country_code":"CN","type":"education","lineage":["https://openalex.org/I90727586"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haiyan Wu","raw_affiliation_strings":["School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China","institution_ids":["https://openalex.org/I90727586"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139747565","display_name":"Chenchen Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I90727586","display_name":"Zhejiang University of Finance and Economics","ror":"https://ror.org/055vj5234","country_code":"CN","type":"education","lineage":["https://openalex.org/I90727586"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chenchen Wang","raw_affiliation_strings":["School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China","institution_ids":["https://openalex.org/I90727586"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139820051","display_name":"Chaoqun Sun","orcid":null},"institutions":[{"id":"https://openalex.org/I90727586","display_name":"Zhejiang University of Finance and Economics","ror":"https://ror.org/055vj5234","country_code":"CN","type":"education","lineage":["https://openalex.org/I90727586"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chaoqun Sun","raw_affiliation_strings":["School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China","institution_ids":["https://openalex.org/I90727586"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139739346","display_name":"Chengxiong Lu","orcid":null},"institutions":[{"id":"https://openalex.org/I90727586","display_name":"Zhejiang University of Finance and Economics","ror":"https://ror.org/055vj5234","country_code":"CN","type":"education","lineage":["https://openalex.org/I90727586"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chengxiong Lu","raw_affiliation_strings":["School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China","institution_ids":["https://openalex.org/I90727586"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139755389","display_name":"Yan-Hong Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I90727586","display_name":"Zhejiang University of Finance and Economics","ror":"https://ror.org/055vj5234","country_code":"CN","type":"education","lineage":["https://openalex.org/I90727586"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yan-Hong Chen","raw_affiliation_strings":["School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China","institution_ids":["https://openalex.org/I90727586"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139728311","display_name":"Zhiqiang Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I90727586","display_name":"Zhejiang University of Finance and Economics","ror":"https://ror.org/055vj5234","country_code":"CN","type":"education","lineage":["https://openalex.org/I90727586"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhiqiang Zhang","raw_affiliation_strings":["School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China","institution_ids":["https://openalex.org/I90727586"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139806223","display_name":"Xiaoqing Feng","orcid":null},"institutions":[{"id":"https://openalex.org/I90727586","display_name":"Zhejiang University of Finance and Economics","ror":"https://ror.org/055vj5234","country_code":"CN","type":"education","lineage":["https://openalex.org/I90727586"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaoqing Feng","raw_affiliation_strings":["School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Information Technology and Artificial Intelligence , Zhejiang University of Finance and Economics , China","institution_ids":["https://openalex.org/I90727586"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I90727586"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.83902543,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"13252","last_page":"13267"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11574","display_name":"Artificial Intelligence in Games","score":0.06870000064373016,"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/T11574","display_name":"Artificial Intelligence in Games","score":0.06870000064373016,"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/T11596","display_name":"Constraint Satisfaction and Optimization","score":0.05829999968409538,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.05779999867081642,"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/process","display_name":"Process (computing)","score":0.3763999938964844},{"id":"https://openalex.org/keywords/reflection","display_name":"Reflection (computer programming)","score":0.33329999446868896},{"id":"https://openalex.org/keywords/perspective","display_name":"Perspective (graphical)","score":0.29580000042915344},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.27459999918937683},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.2745000123977661},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.2709999978542328}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5231999754905701},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.3763999938964844},{"id":"https://openalex.org/C65682993","wikidata":"https://www.wikidata.org/wiki/Q1056451","display_name":"Reflection (computer programming)","level":2,"score":0.33329999446868896},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2980000078678131},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.29580000042915344},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.27459999918937683},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2745000123977661},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.2709999978542328},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.2655999958515167},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.2556000053882599},{"id":"https://openalex.org/C2778348673","wikidata":"https://www.wikidata.org/wiki/Q739302","display_name":"Production (economics)","level":2,"score":0.25189998745918274},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.25029999017715454}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.648","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.648","pdf_url":"https://aclanthology.org/2026.findings-acl.648.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.648","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.648","pdf_url":"https://aclanthology.org/2026.findings-acl.648.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":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166849228.pdf","grobid_xml":"https://content.openalex.org/works/W7166849228.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Retrieval-Augmented":[0],"Generation":[1],"(RAG)":[2],"has":[3],"emerged":[4],"as":[5],"a":[6,66,92,123],"widely":[7],"adopted":[8],"paradigm":[9],"for":[10],"grounding":[11],"Large":[12],"Language":[13],"Models":[14],"(LLMs)":[15],"in":[16],"external":[17],"knowledge.Recent":[18],"agentic":[19,49,152],"RAG":[20,64,148],"systems":[21],"introduce":[22,91],"multi-turn":[23],"reasoning,":[24],"but":[25],"they":[26],"often":[27],"lack":[28],"the":[29,33,72,76,108,129],"capacity":[30],"to":[31,39,54,74,111,163],"evaluate":[32],"utility":[34],"of":[35,78],"retrieved":[36,79],"information,":[37],"leading":[38],"brittle":[40],"reasoning":[41,166],"and":[42,58,160],"suboptimal":[43],"decision-making.We":[44],"propose":[45],"Reflective":[46,63,147],"RAG,":[47],"an":[48],"framework":[50],"that":[51,70,96,146],"incorporates":[52],"self-evaluation":[53],"dynamically":[55],"optimize":[56],"retrieval":[57],"generation":[59],"strategy.At":[60],"its":[61,84,156],"core,":[62],"employs":[65],"reflection":[67,114],"tagging":[68],"mechanism":[69],"allows":[71],"model":[73,109],"critique":[75],"relevance":[77],"content,":[80],"thereby":[81],"explicitly":[82],"guiding":[83],"subsequent":[85],"policy.To":[86],"ensure":[87],"robust":[88],"learning,":[89],"we":[90],"two-stage":[93],"training":[94,158],"procedure":[95],"partially":[97],"decouples":[98],"evaluation":[99],"semantics":[100],"from":[101],"strategy":[102,131],"optimization.First,":[103],"during":[104],"supervised":[105],"fine-tuning":[106],"(SFT),":[107],"learns":[110],"generate":[112],"accurate":[113],"signals":[115],"by":[116,136],"self-correcting":[117],"labels":[118],"based":[119],"on":[120],"internal":[121],"uncertainty.Second,":[122],"reinforcement":[124],"learning":[125],"(RL)":[126],"stage":[127],"optimizes":[128],"agent's":[130],"using":[132],"these":[133],"reflections,":[134],"stabilized":[135],"dynamic":[137],"KL":[138],"regularization.Evaluations":[139],"across":[140],"five":[141],"knowledge-intensive":[142],"QA":[143],"benchmarks":[144],"demonstrate":[145],"consistently":[149],"outperforms":[150],"strong":[151],"baselines.Further":[153],"analysis":[154],"demonstrates":[155],"improved":[157],"stability":[159],"stronger":[161],"generalization":[162],"complex":[164],"multihop":[165],"tasks.":[167]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}