{"id":"https://openalex.org/W4416621886","doi":"https://doi.org/10.1145/3774904.3792191","title":"Plan Then Retrieve: Reinforcement Learning-Guided Complex Reasoning over Knowledge Graphs","display_name":"Plan Then Retrieve: Reinforcement Learning-Guided Complex Reasoning over Knowledge Graphs","publication_year":2026,"publication_date":"2026-04-12","ids":{"openalex":"https://openalex.org/W4416621886","doi":"https://doi.org/10.1145/3774904.3792191"},"language":"en","primary_location":{"id":"doi:10.1145/3774904.3792191","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774904.3792191","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM Web Conference 2026","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3774904.3792191","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5068755935","display_name":"Yanlin Song","orcid":"https://orcid.org/0000-0002-2665-0285"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yanlin Song","raw_affiliation_strings":["School of Computer Science, Wuhan University, Wuhan, China"],"raw_orcid":"https://orcid.org/0009-0009-7617-3630","affiliations":[{"raw_affiliation_string":"School of Computer Science, Wuhan University, Wuhan, China","institution_ids":["https://openalex.org/I37461747"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102827191","display_name":"Ben Liu","orcid":"https://orcid.org/0000-0001-7046-1054"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ben Liu","raw_affiliation_strings":["Ant Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0009-0005-0884-2653","affiliations":[{"raw_affiliation_string":"Ant Group, Hangzhou, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060245641","display_name":"V\u00edctor Guti\u00e9rrez-Basulto","orcid":"https://orcid.org/0000-0002-6117-5459"},"institutions":[{"id":"https://openalex.org/I79510175","display_name":"Cardiff University","ror":"https://ror.org/03kk7td41","country_code":"GB","type":"education","lineage":["https://openalex.org/I79510175"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"V\u00edctor Guti\u00e9rrez-Basulto","raw_affiliation_strings":["School of Computer Science and Informatics, Cardiff University, Cardiff, United Kingdom"],"raw_orcid":"https://orcid.org/0000-0002-6117-5459","affiliations":[{"raw_affiliation_string":"School of Computer Science and Informatics, Cardiff University, Cardiff, United Kingdom","institution_ids":["https://openalex.org/I79510175"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101506606","display_name":"Zhiwei Hu","orcid":"https://orcid.org/0000-0002-8246-4700"},"institutions":[{"id":"https://openalex.org/I119868032","display_name":"Shanxi Agricultural University","ror":"https://ror.org/05e9f5362","country_code":"CN","type":"education","lineage":["https://openalex.org/I119868032"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhiwei Hu","raw_affiliation_strings":["College of Information Science and Engineering, Shanxi Agricultural Unverisity, Taiyuan, China"],"raw_orcid":"https://orcid.org/0000-0002-8246-4700","affiliations":[{"raw_affiliation_string":"College of Information Science and Engineering, Shanxi Agricultural Unverisity, Taiyuan, China","institution_ids":["https://openalex.org/I119868032"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101868563","display_name":"Qianqian Xie","orcid":"https://orcid.org/0000-0002-9588-7454"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qianqian Xie","raw_affiliation_strings":["School of Artificial Intelligence, Wuhan University, Wuhan, China and Center for Language and Information, Wuhan University, Wuhan, China"],"raw_orcid":"https://orcid.org/0000-0002-9588-7454","affiliations":[{"raw_affiliation_string":"School of Artificial Intelligence, Wuhan University, Wuhan, China and Center for Language and Information, Wuhan University, Wuhan, China","institution_ids":["https://openalex.org/I37461747"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102996335","display_name":"Min Peng","orcid":"https://orcid.org/0000-0001-5189-0013"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Min Peng","raw_affiliation_strings":["School of Artificial Intelligence, Wuhan University, Wuhan, China and Center for Language and Information, Wuhan University, Wuhan, China"],"raw_orcid":"https://orcid.org/0000-0002-8766-1105","affiliations":[{"raw_affiliation_string":"School of Artificial Intelligence, Wuhan University, Wuhan, China and Center for Language and Information, Wuhan University, Wuhan, China","institution_ids":["https://openalex.org/I37461747"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077976343","display_name":"Sophia Ananiadou","orcid":"https://orcid.org/0000-0002-4097-9191"},"institutions":[{"id":"https://openalex.org/I28407311","display_name":"University of Manchester","ror":"https://ror.org/027m9bs27","country_code":"GB","type":"education","lineage":["https://openalex.org/I28407311"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Sophia Ananiadou","raw_affiliation_strings":["School of Computer Science, University of Manchester, Manchester, United Kingdom"],"raw_orcid":"https://orcid.org/0000-0002-4097-9191","affiliations":[{"raw_affiliation_string":"School of Computer Science, University of Manchester, Manchester, United Kingdom","institution_ids":["https://openalex.org/I28407311"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5066422711","display_name":"Jeff Z. Pan","orcid":"https://orcid.org/0000-0002-9779-2088"},"institutions":[{"id":"https://openalex.org/I98677209","display_name":"University of Edinburgh","ror":"https://ror.org/01nrxwf90","country_code":"GB","type":"education","lineage":["https://openalex.org/I98677209"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Jeff Z. Pan","raw_affiliation_strings":["ILCC, School of Informatics, University of Edinburgh, Edinburgh, United Kingdom"],"raw_orcid":"https://orcid.org/0000-0002-9779-2088","affiliations":[{"raw_affiliation_string":"ILCC, School of Informatics, University of Edinburgh, Edinburgh, United Kingdom","institution_ids":["https://openalex.org/I98677209"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5068755935"],"corresponding_institution_ids":["https://openalex.org/I37461747"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.01115972,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"3666","last_page":"3676"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.6025999784469604,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.6025999784469604,"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/T10028","display_name":"Topic Modeling","score":0.11339999735355377,"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":0.10819999873638153,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.5863000154495239},{"id":"https://openalex.org/keywords/reasoning-system","display_name":"Reasoning system","score":0.5340999960899353},{"id":"https://openalex.org/keywords/automated-reasoning","display_name":"Automated reasoning","score":0.519599974155426},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.4571000039577484},{"id":"https://openalex.org/keywords/case-based-reasoning","display_name":"Case-based reasoning","score":0.4442000091075897},{"id":"https://openalex.org/keywords/model-based-reasoning","display_name":"Model-based reasoning","score":0.4368000030517578},{"id":"https://openalex.org/keywords/non-monotonic-logic","display_name":"Non-monotonic logic","score":0.429500013589859},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.41920000314712524},{"id":"https://openalex.org/keywords/knowledge-representation-and-reasoning","display_name":"Knowledge representation and reasoning","score":0.40799999237060547},{"id":"https://openalex.org/keywords/opportunistic-reasoning","display_name":"Opportunistic reasoning","score":0.3831000030040741}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7480999827384949},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.5863000154495239},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.5340999960899353},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5267999768257141},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.519599974155426},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.4571000039577484},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.4442000091075897},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.4368000030517578},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.429500013589859},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.41920000314712524},{"id":"https://openalex.org/C161301231","wikidata":"https://www.wikidata.org/wiki/Q3478658","display_name":"Knowledge representation and reasoning","level":2,"score":0.40799999237060547},{"id":"https://openalex.org/C86827895","wikidata":"https://www.wikidata.org/wiki/Q7098582","display_name":"Opportunistic reasoning","level":4,"score":0.3831000030040741},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.38100001215934753},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.3628999888896942},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.3612000048160553},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.34369999170303345},{"id":"https://openalex.org/C2776505523","wikidata":"https://www.wikidata.org/wiki/Q4785468","display_name":"Plan (archaeology)","level":2,"score":0.3416000008583069},{"id":"https://openalex.org/C114073186","wikidata":"https://www.wikidata.org/wiki/Q2631895","display_name":"Automated planning and scheduling","level":2,"score":0.3393000066280365},{"id":"https://openalex.org/C115925183","wikidata":"https://www.wikidata.org/wiki/Q1412694","display_name":"Knowledge-based systems","level":2,"score":0.3375000059604645},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.32179999351501465},{"id":"https://openalex.org/C103057564","wikidata":"https://www.wikidata.org/wiki/Q4751139","display_name":"Analytic reasoning","level":3,"score":0.31709998846054077},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2928999960422516},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.2874000072479248},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.28209999203681946},{"id":"https://openalex.org/C97364631","wikidata":"https://www.wikidata.org/wiki/Q484284","display_name":"Deductive reasoning","level":2,"score":0.2815000116825104},{"id":"https://openalex.org/C84685590","wikidata":"https://www.wikidata.org/wiki/Q1540472","display_name":"Knowledge engineering","level":2,"score":0.27799999713897705},{"id":"https://openalex.org/C2129575","wikidata":"https://www.wikidata.org/wiki/Q54837","display_name":"Semantic Web","level":2,"score":0.2750999927520752},{"id":"https://openalex.org/C83725634","wikidata":"https://www.wikidata.org/wiki/Q7268699","display_name":"Qualitative reasoning","level":2,"score":0.26330000162124634},{"id":"https://openalex.org/C120567893","wikidata":"https://www.wikidata.org/wiki/Q1582085","display_name":"Knowledge extraction","level":2,"score":0.2615000009536743},{"id":"https://openalex.org/C234837","wikidata":"https://www.wikidata.org/wiki/Q1420493","display_name":"Conceptual graph","level":3,"score":0.25529998540878296},{"id":"https://openalex.org/C192028432","wikidata":"https://www.wikidata.org/wiki/Q845739","display_name":"Query language","level":2,"score":0.2540999948978424}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1145/3774904.3792191","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774904.3792191","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM Web Conference 2026","raw_type":"proceedings-article"},{"id":"pmh:oai:https://orca.cardiff.ac.uk:184361","is_oa":false,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4306401195","display_name":"ORCA Online Research @Cardiff (Cardiff University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I79510175","host_organization_name":"Cardiff University","host_organization_lineage":["https://openalex.org/I79510175"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"PeerReviewed"},{"id":"pmh:oai:arXiv.org:2510.20691","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.20691","pdf_url":"https://arxiv.org/pdf/2510.20691","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2510.20691","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.20691","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":"article"}],"best_oa_location":{"id":"doi:10.1145/3774904.3792191","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774904.3792191","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM Web Conference 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Knowledge":[0],"Graph":[1],"Question":[2],"Answering":[3],"(KGQA)":[4],"aims":[5],"to":[6,34,36,65,79,85,110,178,188,213,218],"answer":[7],"natural":[8],"language":[9,20],"questions":[10,181],"by":[11],"reasoning":[12,29,48,74,86,141],"over":[13,236],"structured":[14,140],"knowledge":[15,42,91,125],"graphs":[16],"(KGs).":[17],"While":[18],"large":[19],"models":[21],"(LLMs)":[22],"have":[23],"advanced":[24],"KGQA":[25,101,228],"through":[26],"their":[27,73],"strong":[28,237],"capabilities,":[30],"existing":[31],"methods":[32],"continue":[33],"struggle":[35],"fully":[37],"exploit":[38],"both":[39],"the":[40,47,144,211],"rich":[41],"encoded":[43],"in":[44,53],"KGs":[45],"and":[46,62,72,114,120,142,161,185,207,216,221,244,252],"capabilities":[49],"of":[50],"LLMs,":[51],"particularly":[52],"complex":[54,180,247],"scenarios.":[55],"They":[56],"often":[57],"assume":[58],"complete":[59],"KG":[60,119,220],"coverage":[61],"lack":[63],"mechanisms":[64],"judge":[66],"when":[67,89,215],"external":[68],"information":[69],"is":[70,200],"needed,":[71],"remains":[75],"locally":[76],"myopic,":[77],"failing":[78],"maintain":[80],"coherent":[81],"multi-step":[82,195],"planning,":[83],"leading":[84],"failures":[87],"even":[88,239],"relevant":[90],"exists.":[92],"We":[93],"propose":[94],"Graph-RFT,":[95],"a":[96,104,129,135,151,165,174,203],"novel":[97,152],"two-stage":[98],"reinforcement":[99,155],"fine-tuning":[100,132],"framework":[102],"with":[103,134,164,202,240],"''plan\u2013KGsearch\u2013and\u2013Websearch\u2013during\u2013think''":[105],"paradigm,":[106],"that":[107,231],"enables":[108],"LLMs":[109],"perform":[111],"autonomous":[112],"planning":[113,160,176],"adaptive":[115],"retrieval":[116,162,170,223],"scheduling":[117],"across":[118],"web":[121,222],"sources":[122],"under":[123],"incomplete":[124],"conditions.":[126],"Graph-RFT":[127,232],"introduces":[128,150],"chain-of-thought":[130],"(CoT)":[131],"method":[133],"customized":[136],"plan\u2013retrieval":[137,153],"dataset":[138],"activates":[139],"resolves":[143],"GRPO":[145],"cold-start":[146],"problem.":[147],"It":[148,172],"then":[149],"guided":[154],"learning":[156],"process":[157,199],"integrates":[158],"explicit":[159],"actions":[163],"multi-reward":[166,204],"design,":[167],"enabling":[168,210],"coverage-aware":[169],"scheduling.":[171],"employs":[173],"Cartesian-inspired":[175],"module":[177],"decompose":[179],"into":[182],"ordered":[183],"sub-questions,":[184],"logical":[186],"expression":[187],"guide":[189],"tool":[190,253],"invocation":[191],"for":[192],"globally":[193],"consistent":[194],"reasoning.":[196],"This":[197],"reasoning\u2013retrieval":[198],"optimized":[201],"combining":[205],"outcome":[206],"retrieval-specific":[208],"signals,":[209],"model":[212],"learn":[214],"how":[217],"combine":[219],"effectively.":[224],"Experiments":[225],"on":[226],"multiple":[227],"benchmarks":[229],"demonstrate":[230],"achieves":[233],"superior":[234],"performance":[235],"baselines,":[238],"smaller":[241],"LLM":[242],"backbones,":[243],"substantially":[245],"improves":[246],"question":[248],"decomposition,":[249],"factual":[250],"coverage,":[251],"coordination.":[254]},"counts_by_year":[],"updated_date":"2026-04-29T06:10:49.150238","created_date":"2025-10-25T00:00:00"}