{"id":"https://openalex.org/W7213372913","doi":null,"title":"GraMRAG: Orchestrating Multi-Agent Multi-Step Reasoning via Graph Memory with Reinforcement Learning","display_name":"GraMRAG: Orchestrating Multi-Agent Multi-Step Reasoning via Graph Memory with Reinforcement Learning","publication_year":2026,"publication_date":"2026-09-12","ids":{"openalex":"https://openalex.org/W7213372913"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.14066","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.14066","pdf_url":"https://arxiv.org/pdf/2609.14066","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.14066","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139816454","display_name":"Zhongyu Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Wang, Zhongyu","raw_affiliation_strings":["Beihang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beihang University","institution_ids":["https://openalex.org/I82880672"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5139816454"],"corresponding_institution_ids":["https://openalex.org/I82880672"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9994999766349792,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9994999766349792,"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/T10456","display_name":"Multi-Agent Systems and Negotiation","score":9.999999747378752e-05,"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":9.999999747378752e-05,"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/graph-memory","display_name":"graph memory","score":0.8069999814033508},{"id":"https://openalex.org/keywords/multi-step-reasoning","display_name":"multi-step reasoning","score":0.746999979019165},{"id":"https://openalex.org/keywords/multimodal-reasoning","display_name":"multimodal reasoning","score":0.8190000057220459},{"id":"https://openalex.org/keywords/directed-acyclic-graphs","display_name":"directed acyclic graphs","score":0.7990000247955322},{"id":"https://openalex.org/keywords/react","display_name":"React","score":0.48100000619888306},{"id":"https://openalex.org/keywords/credit-assignment","display_name":"credit assignment","score":0.9520000219345093},{"id":"https://openalex.org/keywords/long-horizon-reasoning","display_name":"long-horizon reasoning","score":0.9150000214576721}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7773000001907349},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5824999809265137},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5072000026702881},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.45739999413490295},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.4449999928474426},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.41760000586509705},{"id":"https://openalex.org/C2780929884","wikidata":"https://www.wikidata.org/wiki/Q737460","display_name":"Blindness","level":2,"score":0.397599995136261},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.39559999108314514},{"id":"https://openalex.org/C86827895","wikidata":"https://www.wikidata.org/wiki/Q7098582","display_name":"Opportunistic reasoning","level":4,"score":0.38960000872612},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.3824000060558319},{"id":"https://openalex.org/C146380142","wikidata":"https://www.wikidata.org/wiki/Q1137726","display_name":"Directed graph","level":2,"score":0.3668999969959259},{"id":"https://openalex.org/C82687282","wikidata":"https://www.wikidata.org/wiki/Q66221","display_name":"Auxiliary memory","level":2,"score":0.3402000069618225},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.3377000093460083},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.335999995470047},{"id":"https://openalex.org/C115086926","wikidata":"https://www.wikidata.org/wiki/Q17004651","display_name":"Causal reasoning","level":3,"score":0.3273000121116638},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.3091999888420105},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.30869999527931213},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.2971000075340271},{"id":"https://openalex.org/C161301231","wikidata":"https://www.wikidata.org/wiki/Q3478658","display_name":"Knowledge representation and reasoning","level":2,"score":0.2883000075817108},{"id":"https://openalex.org/C83725634","wikidata":"https://www.wikidata.org/wiki/Q7268699","display_name":"Qualitative reasoning","level":2,"score":0.28299999237060547},{"id":"https://openalex.org/C106891557","wikidata":"https://www.wikidata.org/wiki/Q4148051","display_name":"Wait-for graph","level":4,"score":0.2824999988079071},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.28139999508857727},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.26179999113082886},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.25589999556541443},{"id":"https://openalex.org/C97364631","wikidata":"https://www.wikidata.org/wiki/Q484284","display_name":"Deductive reasoning","level":2,"score":0.25049999356269836},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:arXiv.org:2609.14066","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.14066","pdf_url":"https://arxiv.org/pdf/2609.14066","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.14066","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.14066","pdf_url":"https://arxiv.org/pdf/2609.14066","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7213372913.pdf","grobid_xml":"https://content.openalex.org/works/W7213372913.grobid-xml"},"referenced_works_count":17,"referenced_works":["https://openalex.org/W4304195432","https://openalex.org/W4391631327","https://openalex.org/W4407759090","https://openalex.org/W4414939355","https://openalex.org/W4414990013","https://openalex.org/W4415272886","https://openalex.org/W4415276218","https://openalex.org/W4416679470","https://openalex.org/W4416780398","https://openalex.org/W4417095241","https://openalex.org/W4417299405","https://openalex.org/W6929370958","https://openalex.org/W6947947872","https://openalex.org/W7102388436","https://openalex.org/W7102411227","https://openalex.org/W7126150514","https://openalex.org/W7129042046"],"related_works":["https://openalex.org/W7152019015","https://openalex.org/W7129042046","https://openalex.org/W7214724280","https://openalex.org/W7204889407","https://openalex.org/W7129099921","https://openalex.org/W7117339518","https://openalex.org/W7171972818","https://openalex.org/W4415539809","https://openalex.org/W7140295500","https://openalex.org/W7168857248"],"abstract_inverted_index":{"Although":[0],"existing":[1,154],"multi-agent":[2,46],"Retrieval-Augmented":[3],"Generation":[4],"(RAG)":[5],"systems":[6],"have":[7],"demonstrated":[8],"promise":[9],"on":[10,144,160],"complex":[11,161],"multimodal":[12,53,60,89,146],"reasoning":[13,20,67,83,95,140,163],"tasks,":[14],"they":[15],"remain":[16],"fundamentally":[17],"limited":[18],"in":[19],"depth":[21],"and":[22,29,111,130,156],"memory":[23,54,90],"structure,":[24],"suffering":[25],"from":[26],"inadequate":[27],"retrieval":[28],"state":[30,109],"blindness":[31,110],"when":[32],"answering":[33],"knowledge-intensive":[34],"questions.":[35],"To":[36],"address":[37],"these":[38],"limitations,":[39],"we":[40,116],"propose":[41,117],"GraMRAG,":[42],"a":[43,51,64,75,88,97],"graph":[44,55,91,101,124],"memory-guided":[45],"RAG":[47],"framework":[48],"that":[49,69,92,122,149],"integrates":[50],"dynamic":[52,98],"to":[56,107],"enable":[57],"stable,":[58],"multi-step":[59,139],"reasoning.":[61],"We":[62,85],"introduce":[63],"vision-text":[65],"bridged":[66],"paradigm":[68],"unifies":[70],"multi-scale":[71],"entity":[72],"cropping":[73],"with":[74],"ReAct-style":[76],"visual":[77],"toolchain,":[78],"enhancing":[79],"the":[80],"long-horizon":[81,162],"cross-modal":[82],"capability.":[84],"further":[86],"construct":[87],"formalizes":[93],"agent":[94],"as":[96],"directed":[99],"acyclic":[100],"(DAG),":[102],"explicitly":[103],"modeling":[104],"action-observation":[105],"dependencies":[106],"mitigate":[108],"suppress":[112],"redundant":[113],"retrieval.":[114],"Moreover,":[115],"Topology-Aware":[118],"Policy":[119],"Optimization":[120],"(TAPO)":[121],"leverages":[123],"topology":[125],"for":[126],"critical":[127],"path":[128],"identification":[129],"targeted":[131],"node":[132],"pruning,":[133],"enabling":[134],"fine-grained":[135],"credit":[136],"assignment":[137],"across":[138],"trajectories.":[141],"Extensive":[142],"experiments":[143],"challenging":[145],"benchmarks":[147],"demonstrate":[148],"our":[150],"approach":[151],"consistently":[152],"outperforms":[153],"baselines":[155],"achieves":[157],"state-of-the-art":[158],"performance":[159],"tasks.":[164]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-16T00:00:00"}