{"id":"https://openalex.org/W7127638641","doi":"https://doi.org/10.48550/arxiv.2602.02518","title":"GraphDancer: Training LLMs to Explore and Reason over Graphs via Two-Stage Curriculum Post-Training","display_name":"GraphDancer: Training LLMs to Explore and Reason over Graphs via Two-Stage Curriculum Post-Training","publication_year":2026,"publication_date":"2026-01-24","ids":{"openalex":"https://openalex.org/W7127638641","doi":"https://doi.org/10.48550/arxiv.2602.02518"},"language":null,"primary_location":{"id":"pmh:oai:arXiv.org:2602.02518","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2602.02518","pdf_url":"https://arxiv.org/pdf/2602.02518","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2602.02518","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100624468","display_name":"Yuyang Bai","orcid":"https://orcid.org/0009-0005-4363-5350","observed_orcids":["https://orcid.org/0009-0005-4363-5350"]},"institutions":[{"id":"https://openalex.org/I91045830","display_name":"Texas A&M University","ror":"https://ror.org/01f5ytq51","country_code":"US","type":"education","lineage":["https://openalex.org/I91045830"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bai, Yuyang","raw_affiliation_strings":["Texas A&M University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Texas A&M University ,","institution_ids":["https://openalex.org/I91045830"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125004748","display_name":"Zhuofeng Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I91045830","display_name":"Texas A&M University","ror":"https://ror.org/01f5ytq51","country_code":"US","type":"education","lineage":["https://openalex.org/I91045830"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Li, Zhuofeng","raw_affiliation_strings":["Texas A&M University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Texas A&M University ,","institution_ids":["https://openalex.org/I91045830"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140766233","display_name":"Ping Nie","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I151746483","display_name":"University of Waterloo","ror":"https://ror.org/01aff2v68","country_code":"CA","type":"education","lineage":["https://openalex.org/I151746483"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Nie, Ping","raw_affiliation_strings":["University of Waterloo ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Waterloo ,","institution_ids":["https://openalex.org/I151746483"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125020225","display_name":"Jianwen Xie","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210086225","display_name":"Lambda Instruments (United States)","ror":"https://ror.org/00thhxf09","country_code":"US","type":"company","lineage":["https://openalex.org/I4210086225"]},{"id":"https://openalex.org/I4210160406","display_name":"Lambda Science (United States)","ror":"https://ror.org/04wae4s50","country_code":"US","type":"company","lineage":["https://openalex.org/I4210160406"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xie, Jianwen","raw_affiliation_strings":["Lambda ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Lambda ,","institution_ids":["https://openalex.org/I4210086225","https://openalex.org/I4210160406"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5125082038","display_name":"Yu Zhang","orcid":"https://orcid.org/0000-0002-8807-6389","observed_orcids":["https://orcid.org/0000-0002-8807-6389"]},"institutions":[{"id":"https://openalex.org/I91045830","display_name":"Texas A&M University","ror":"https://ror.org/01f5ytq51","country_code":"US","type":"education","lineage":["https://openalex.org/I91045830"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhang, Yu","raw_affiliation_strings":["Texas A&M University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Texas A&M University ,","institution_ids":["https://openalex.org/I91045830"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"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/T10028","display_name":"Topic Modeling","score":0.8661999702453613,"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/T10028","display_name":"Topic Modeling","score":0.8661999702453613,"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":0.05389999970793724,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.02370000071823597,"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/large-language-models","display_name":"large language models","score":0.8510000109672546},{"id":"https://openalex.org/keywords/graph-reasoning","display_name":"graph reasoning","score":0.8690000176429749},{"id":"https://openalex.org/keywords/heterogeneous-graphs","display_name":"heterogeneous graphs","score":0.9419999718666077},{"id":"https://openalex.org/keywords/post-training","display_name":"post-training","score":0.6389999985694885},{"id":"https://openalex.org/keywords/curriculum-learning","display_name":"curriculum learning","score":0.9150000214576721},{"id":"https://openalex.org/keywords/function-calling","display_name":"function calling","score":0.5920000076293945},{"id":"https://openalex.org/keywords/domain-generalization","display_name":"domain generalization","score":0.8730000257492065},{"id":"https://openalex.org/keywords/out-of-distribution-generalization","display_name":"out-of-distribution generalization","score":0.8080000281333923}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6269999742507935},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.5677000284194946},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5597000122070312},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5318999886512756},{"id":"https://openalex.org/C28034677","wikidata":"https://www.wikidata.org/wiki/Q17092530","display_name":"Interleaving","level":2,"score":0.5058000087738037},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.49939998984336853},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4587000012397766},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.44670000672340393},{"id":"https://openalex.org/C47177190","wikidata":"https://www.wikidata.org/wiki/Q207137","display_name":"Curriculum","level":2,"score":0.4449000060558319},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.40230000019073486},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.3889000117778778},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33550000190734863},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.32850000262260437},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.3199999928474426},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.29670000076293945},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.2874999940395355},{"id":"https://openalex.org/C143587482","wikidata":"https://www.wikidata.org/wiki/Q1543216","display_name":"Iterative and incremental development","level":2,"score":0.26330000162124634},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C2777152325","wikidata":"https://www.wikidata.org/wiki/Q108163","display_name":"Proposition","level":2,"score":0.25839999318122864}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2602.02518","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2602.02518","pdf_url":"https://arxiv.org/pdf/2602.02518","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2602.02518","doi":"https://doi.org/10.48550/arxiv.2602.02518","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.02518","pdf_url":null,"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2602.02518","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2602.02518","pdf_url":"https://arxiv.org/pdf/2602.02518","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[{"score":0.8640525341033936,"display_name":"Quality education","id":"https://metadata.un.org/sdg/4"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7127638641.pdf","grobid_xml":"https://content.openalex.org/works/W7127638641.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2,31],"(LLMs)":[3],"increasingly":[4],"rely":[5],"on":[6,130,136,142],"external":[7],"knowledge":[8,15],"to":[9,32,41,59,78,93],"improve":[10],"factuality,":[11],"yet":[12],"many":[13],"real-world":[14],"sources":[16],"are":[17],"organized":[18],"as":[19],"heterogeneous":[20],"graphs":[21,29,62],"rather":[22],"than":[23],"plain":[24],"text.":[25],"Reasoning":[26],"over":[27,61],"such":[28],"requires":[30],"follow":[33],"schema-defined":[34],"relations":[35],"through":[36],"precise":[37],"function":[38,69],"calls":[39],"and":[40,97,140,145,169],"aggregate":[42],"evidence":[43],"across":[44],"multiple":[45],"rounds":[46],"of":[47,104,118,166],"interaction.":[48],"We":[49,127],"propose":[50],"GraphDancer,":[51],"a":[52,107,131,152],"two-stage":[53],"post-training":[54],"framework":[55],"that":[56,110],"teaches":[57,74,91],"LLMs":[58],"reason":[60],"by":[63,114,134],"interleaving":[64],"natural-language":[65],"reasoning":[66,170],"with":[67,80,159],"graph":[68,82,167],"execution.":[70],"The":[71,101],"first":[72],"stage":[73,89],"the":[75,81,87,115],"model":[76],"how":[77],"interact":[79],"under":[83],"rule-based":[84],"rewards,":[85],"while":[86],"second":[88],"further":[90],"it":[92],"prefer":[94],"more":[95],"grounded":[96],"efficient":[98],"interaction":[99],"trajectories.":[100],"key":[102],"novelty":[103],"GraphDancer":[105,129,155],"is":[106],"graph-aware":[108],"curriculum":[109],"organizes":[111],"both":[112],"stages":[113],"structural":[116],"complexity":[117],"information-seeking":[119],"trajectories,":[120],"progressively":[121],"increasing":[122],"task":[123],"difficulty":[124],"during":[125],"training.":[126],"evaluate":[128],"multi-domain":[132],"benchmark":[133],"training":[135],"one":[137],"domain":[138],"only":[139,151],"testing":[141],"unseen":[143],"domains":[144],"out-of-distribution":[146],"question":[147],"types.":[148],"Despite":[149],"using":[150],"3B":[153],"backbone,":[154],"outperforms":[156],"baselines":[157],"equipped":[158],"larger/stronger":[160],"backbones,":[161],"demonstrating":[162],"robust":[163],"cross-domain":[164],"generalization":[165],"exploration":[168],"skills.":[171],"Our":[172],"code":[173],"can":[174],"be":[175],"found":[176],"at":[177],"https://github.com/leopoldwhite/GraphDancer.":[178]},"counts_by_year":[],"updated_date":"2026-09-30T09:27:27.532063","created_date":"2026-02-06T00:00:00"}