{"id":"https://openalex.org/W7168399467","doi":"https://doi.org/10.1145/3805712.3809649","title":"Exploration-and-Thinking: Agentic Reasoning over Knowledge Graphs via an LLM-RL Synergized Framework","display_name":"Exploration-and-Thinking: Agentic Reasoning over Knowledge Graphs via an LLM-RL Synergized Framework","publication_year":2026,"publication_date":"2026-07-15","ids":{"openalex":"https://openalex.org/W7168399467","doi":"https://doi.org/10.1145/3805712.3809649"},"language":null,"primary_location":{"id":"doi:10.1145/3805712.3809649","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805712.3809649","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 49th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3805712.3809649","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5140913898","display_name":"Yi Xia","orcid":"https://orcid.org/0000-0003-0048-9929"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yi Xia","raw_affiliation_strings":["State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China"],"raw_orcid":"https://orcid.org/0000-0003-0048-9929","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140944347","display_name":"Gang Zhou","orcid":"https://orcid.org/0009-0007-4180-0986"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gang Zhou","raw_affiliation_strings":["State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China"],"raw_orcid":"https://orcid.org/0009-0007-4180-0986","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140806068","display_name":"Jing Chen","orcid":"https://orcid.org/0009-0002-9433-1398"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jing Chen","raw_affiliation_strings":["State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China"],"raw_orcid":"https://orcid.org/0009-0002-9433-1398","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140851784","display_name":"Xiaohui Chen","orcid":"https://orcid.org/0000-0001-8509-6790"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaohui Chen","raw_affiliation_strings":["State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China"],"raw_orcid":"https://orcid.org/0000-0001-8509-6790","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140873506","display_name":"Qinlong Fan","orcid":"https://orcid.org/0009-0005-6147-399X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qinlong Fan","raw_affiliation_strings":["State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China"],"raw_orcid":"https://orcid.org/0009-0005-6147-399X","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5140906897","display_name":"Shunhang Li","orcid":"https://orcid.org/0009-0005-5942-0928"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shunhang Li","raw_affiliation_strings":["State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China"],"raw_orcid":"https://orcid.org/0009-0005-5942-0928","affiliations":[{"raw_affiliation_string":"State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou, Henan, China","institution_ids":[]}]}],"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":"2049","last_page":"2060"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":null,"topics":[],"keywords":[{"id":"https://openalex.org/keywords/knowledge-representation-and-reasoning","display_name":"Knowledge representation and reasoning","score":0.40070000290870667},{"id":"https://openalex.org/keywords/knowledge-graph","display_name":"Knowledge graph","score":0.3206000030040741},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.3199000060558319},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.29789999127388},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.2808000147342682},{"id":"https://openalex.org/keywords/agency","display_name":"Agency (philosophy)","score":0.26930001378059387}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6157000064849854},{"id":"https://openalex.org/C161301231","wikidata":"https://www.wikidata.org/wiki/Q3478658","display_name":"Knowledge representation and reasoning","level":2,"score":0.40070000290870667},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3921000063419342},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.3732999861240387},{"id":"https://openalex.org/C2987255567","wikidata":"https://www.wikidata.org/wiki/Q33002955","display_name":"Knowledge graph","level":2,"score":0.3206000030040741},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.3199000060558319},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3003999888896942},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.29789999127388},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2808000147342682},{"id":"https://openalex.org/C108170787","wikidata":"https://www.wikidata.org/wiki/Q3951828","display_name":"Agency (philosophy)","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.26809999346733093},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.2535000145435333},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2529999911785126},{"id":"https://openalex.org/C115925183","wikidata":"https://www.wikidata.org/wiki/Q1412694","display_name":"Knowledge-based systems","level":2,"score":0.25049999356269836}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3805712.3809649","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805712.3809649","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 49th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3805712.3809649","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805712.3809649","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 49th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W2107306718","https://openalex.org/W2511149293","https://openalex.org/W2889344053","https://openalex.org/W2964120615","https://openalex.org/W3098682001","https://openalex.org/W3100058503","https://openalex.org/W3156636935","https://openalex.org/W4292635665","https://openalex.org/W4384918448","https://openalex.org/W4385567039","https://openalex.org/W4385569771","https://openalex.org/W4385571689","https://openalex.org/W4392781917","https://openalex.org/W4393160905","https://openalex.org/W4399253046","https://openalex.org/W4399401010","https://openalex.org/W4400525003","https://openalex.org/W4400526005","https://openalex.org/W4402670441","https://openalex.org/W4403791927","https://openalex.org/W4404347859","https://openalex.org/W4409657242","https://openalex.org/W4409670780","https://openalex.org/W4412376966","https://openalex.org/W4412377026","https://openalex.org/W7131898026"],"related_works":[],"abstract_inverted_index":{"While":[0],"Knowledge":[1],"Graphs":[2],"(KGs)":[3],"can":[4],"ground":[5],"Large":[6],"Language":[7],"Models":[8],"(LLMs)":[9],"in":[10,117],"factual":[11],"knowledge,":[12],"existing":[13],"LLM-KG":[14],"integration":[15],"methods":[16],"for":[17,70],"complex":[18],"reasoning":[19,43,74,139],"are":[20,80],"plagued":[21],"by":[22],"computational":[23],"inefficiency":[24],"and":[25,34,72,98,111,127],"semantic":[26],"inconsistency.":[27],"The":[28],"tight":[29],"coupling":[30],"of":[31,141],"LLM":[32],"inference":[33],"KG":[35],"traversal":[36],"leads":[37],"to":[38],"prohibitive":[39],"costs,":[40],"while":[41,143],"spurious":[42],"paths":[44,110],"often":[45],"misguide":[46],"learning-based":[47],"agents,":[48],"causing":[49],"reward":[50,102],"hacking.":[51],"To":[52],"this":[53],"end,":[54],"we":[55],"propose":[56],"EAT":[57,131],"(Exploration-and-Thinking),":[58],"a":[59,145],"novel":[60],"agentic":[61],"framework":[62],"that":[63,88,105,130],"synergizes":[64],"LLMs":[65],"with":[66],"Reinforcement":[67],"Learning":[68],"(RL)":[69],"efficient":[71],"faithful":[73],"on":[75,122],"KGs.":[76],"EAT's":[77],"core":[78],"innovations":[79],"twofold:":[81],"(1)":[82],"an":[83,100],"adaptive":[84],"retrieval-augmented":[85],"generation":[86],"mechanism":[87],"decouples":[89],"language":[90],"comprehension":[91],"from":[92],"structured":[93],"exploration,":[94],"dramatically":[95],"improving":[96],"efficiency;":[97],"(2)":[99],"LLM-guided":[101],"shaping":[103],"strategy":[104],"explicitly":[106],"penalizes":[107],"semantically":[108],"inconsistent":[109],"promotes":[112],"logically":[113],"valid":[114],"trajectories":[115],"grounded":[116],"the":[118,138],"KG.":[119],"Extensive":[120],"experiments":[121],"benchmarks":[123],"like":[124],"WebQSP,":[125],"CWQ,":[126],"GrailQA":[128],"show":[129],"achieves":[132],"state-of-the-art":[133],"performance.":[134],"Notably,":[135],"it":[136],"surpasses":[137],"capability":[140],"GPT-4o":[142],"utilizing":[144],"significantly":[146],"smaller":[147],"8B-parameter":[148],"LLM.":[149]},"counts_by_year":[],"updated_date":"2026-07-17T05:52:16.776730","created_date":"2026-07-16T00:00:00"}