{"id":"https://openalex.org/W7163019235","doi":"https://doi.org/10.48550/arxiv.2605.30786","title":"AbstainGNN: Teaching Graph Neural Networks to Abstain for Graph Classification","display_name":"AbstainGNN: Teaching Graph Neural Networks to Abstain for Graph Classification","publication_year":2026,"publication_date":"2026-05-29","ids":{"openalex":"https://openalex.org/W7163019235","doi":"https://doi.org/10.48550/arxiv.2605.30786"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.30786","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.30786","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.30786","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101641900","display_name":"Xixun Lin","orcid":"https://orcid.org/0009-0004-6645-0597"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Xixun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137596665","display_name":"Zhiheng Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Zhiheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137556083","display_name":"Zhengyin Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Zhengyin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137542765","display_name":"Yancheng Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yancheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137528528","display_name":"Shuai Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Shuai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137569958","display_name":"Ge Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Ge","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137571370","display_name":"Shichao Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Shichao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089307887","display_name":"Lixin Zou","orcid":"https://orcid.org/0000-0001-6755-871X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zou, Lixin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137569946","display_name":"Chuan Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Chuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137599294","display_name":"Peng Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Peng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137541735","display_name":"Shirui Pan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pan, Shirui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137601827","display_name":"Yanan Cao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cao, Yanan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"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":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9824000000953674,"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":0.9824000000953674,"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/T12292","display_name":"Graph Theory and Algorithms","score":0.0032999999821186066,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.0027000000700354576,"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","display_name":"Graph","score":0.5687999725341797},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.47440001368522644},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4251999855041504},{"id":"https://openalex.org/keywords/heuristic","display_name":"Heuristic","score":0.3801000118255615},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.33660000562667847}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6942999958992004},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5687999725341797},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5559999942779541},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5174999833106995},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.47440001368522644},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4251999855041504},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.3801000118255615},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.33660000562667847},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.3310999870300293},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.32600000500679016},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2671000063419342},{"id":"https://openalex.org/C88230418","wikidata":"https://www.wikidata.org/wiki/Q131476","display_name":"Graph theory","level":2,"score":0.2619999945163727},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.25929999351501465}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.30786","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.30786","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":"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":"doi:10.48550/arxiv.2605.30786","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.30786","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Graph":[0],"classification":[1,75,122,178],"is":[2],"a":[3,68,128,134],"core":[4],"task":[5],"in":[6,16,47,57],"graph":[7,17,28,74,107],"data":[8],"mining":[9],"with":[10,76],"widespread":[11],"real-world":[12],"applications.":[13],"Recent":[14],"advances":[15],"neural":[18],"networks":[19],"(GNNs)":[20],"have":[21],"led":[22],"to":[23,36,81],"substantial":[24],"performance":[25,179],"improvements":[26],"for":[27,73,103,138],"classification.":[29],"However,":[30],"existing":[31,112,173],"GNNs":[32,80],"are":[33],"typically":[34],"forced":[35],"make":[37],"predictions":[38,84],"even":[39],"under":[40,180],"high":[41],"uncertainty":[42],"or":[43],"unknown":[44],"conditions,":[45],"resulting":[46],"unreliable":[48],"decisions":[49],"that":[50,170],"can":[51],"severely":[52],"impact":[53],"downstream":[54],"tasks,":[55],"particularly":[56],"safety-critical":[58],"scenarios.":[59],"To":[60],"address":[61],"this":[62,143],"critical":[63],"limitation,":[64],"we":[65,116,146],"propose":[66],"AbstainGNN,":[67],"novel":[69],"and":[70,98,124,132,159],"theory-driven":[71],"framework":[72],"abstention,":[77],"which":[78],"enables":[79],"reject":[82],"uncertain":[83],"instead":[85],"of":[86,106,155],"producing":[87],"incorrect":[88],"decisions.":[89],"Specifically,":[90],"AbstainGNN":[91,171],"explicitly":[92],"models":[93],"both":[94],"the":[95,99,119,181],"predictive":[96,156],"function":[97,157,161],"abstention":[100,114,160,174],"function,":[101],"allowing":[102],"effective":[104],"utilization":[105],"structural":[108],"information.":[109],"Moreover,":[110],"unlike":[111],"heuristic":[113],"methods,":[115,175],"theoretically":[117],"characterize":[118],"trade-off":[120],"between":[121],"errors":[123],"rejection":[125,183],"costs":[126],"from":[127],"PAC-Bayesian":[129],"generalization":[130],"perspective,":[131],"derive":[133],"unified":[135],"learning":[136],"objective":[137],"model":[139],"optimization.":[140],"Guided":[141],"by":[142],"theoretical":[144],"insight,":[145],"further":[147],"develop":[148],"an":[149],"efficient":[150],"two-stage":[151],"training":[152],"strategy":[153],"consisting":[154],"warm-start":[158],"calibration.":[162],"Extensive":[163],"experiments":[164],"on":[165],"five":[166],"benchmark":[167],"datasets":[168],"show":[169],"outperforms":[172],"achieving":[176],"superior":[177],"same":[182],"rates.":[184]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-02T00:00:00"}