{"id":"https://openalex.org/W7160598187","doi":"https://doi.org/10.48550/arxiv.2605.05463","title":"Robustness of Graph Self-Supervised Learning to Real-World Noise: A Case Study on Text-Driven Biomedical Graphs","display_name":"Robustness of Graph Self-Supervised Learning to Real-World Noise: A Case Study on Text-Driven Biomedical Graphs","publication_year":2026,"publication_date":"2026-05-06","ids":{"openalex":"https://openalex.org/W7160598187","doi":"https://doi.org/10.48550/arxiv.2605.05463"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.05463","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05463","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,"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.2605.05463","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5107490449","display_name":"Othmane Kabal","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kabal, Othmane","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103913523","display_name":"Mounira Harzallah","orcid":"https://orcid.org/0000-0002-9473-427X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Harzallah, Mounira","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021901258","display_name":"Fabrice Guillet","orcid":"https://orcid.org/0000-0002-1397-6497"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guillet, Fabrice","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135722689","display_name":"Hideaki Takeda","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Takeda, Hideaki","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5081854769","display_name":"Ryutaro Ichise","orcid":"https://orcid.org/0000-0001-8474-0150"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ichise, Ryutaro","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":0.8986999988555908,"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":0.8986999988555908,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.039000000804662704,"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":0.03189999982714653,"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/robustness","display_name":"Robustness (evolution)","score":0.5934000015258789},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.4650000035762787},{"id":"https://openalex.org/keywords/feature-learning","display_name":"Feature learning","score":0.40380001068115234},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3352999985218048},{"id":"https://openalex.org/keywords/feature-engineering","display_name":"Feature engineering","score":0.3345000147819519},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.3160000145435333}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.753600001335144},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5934000015258789},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.5220000147819519},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4797999858856201},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4650000035762787},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.46059998869895935},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.40380001068115234},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3352999985218048},{"id":"https://openalex.org/C2778827112","wikidata":"https://www.wikidata.org/wiki/Q22245680","display_name":"Feature engineering","level":3,"score":0.3345000147819519},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3160000145435333},{"id":"https://openalex.org/C100595998","wikidata":"https://www.wikidata.org/wiki/Q11731931","display_name":"Graph kernel","level":5,"score":0.30469998717308044},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.29829999804496765},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.26820001006126404},{"id":"https://openalex.org/C177877439","wikidata":"https://www.wikidata.org/wiki/Q7604413","display_name":"Statistical relational learning","level":3,"score":0.26179999113082886},{"id":"https://openalex.org/C2987255567","wikidata":"https://www.wikidata.org/wiki/Q33002955","display_name":"Knowledge graph","level":2,"score":0.26109999418258667},{"id":"https://openalex.org/C88230418","wikidata":"https://www.wikidata.org/wiki/Q131476","display_name":"Graph theory","level":2,"score":0.2522999942302704},{"id":"https://openalex.org/C157406716","wikidata":"https://www.wikidata.org/wiki/Q4115842","display_name":"Topological graph theory","level":5,"score":0.2515000104904175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.05463","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05463","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,"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.2605.05463","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.05463","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,"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":[{"score":0.414751797914505,"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Graph":[0,87,143],"Self-Supervised":[1],"Learning":[2],"(GSSL)":[3],"offers":[4],"a":[5,90,105,110,117,128,193,232],"powerful":[6],"paradigm":[7],"for":[8,40,79,221],"learning":[9],"graph":[10,96,98,112],"representations":[11],"without":[12],"labeled":[13],"data.":[14],"However,":[15],"existing":[16],"work":[17],"assumes":[18],"clean,":[19],"manually":[20],"curated":[21],"graphs.":[22,215],"Recent":[23],"advances":[24],"in":[25,136],"NLP":[26],"enable":[27],"the":[28,69],"large-scale":[29],"automatic":[30,95],"extraction":[31],"of":[32,49,73],"knowledge":[33],"graphs":[34,78,227],"from":[35,114,157],"text,":[36],"opening":[37],"new":[38],"opportunities":[39],"GSSL":[41,74,88,223],"while":[42,207],"introducing":[43],"substantial":[44],"real-world":[45],"noise.":[46],"This":[47],"type":[48],"noise":[50,154,180],"remains":[51],"largely":[52],"unexplored,":[53],"as":[54],"prior":[55],"robustness":[56,137],"studies":[57],"typically":[58],"rely":[59],"on":[60,76,184,213],"synthetic":[61],"perturbations.":[62],"To":[63],"address":[64],"this":[65],"gap,":[66],"we":[67],"present":[68],"first":[70],"comprehensive":[71],"evaluation":[72,103],"methods":[75],"text-driven":[77,205],"unsupervised":[80],"term":[81],"typing.":[82],"We":[83],"introduce":[84],"Noise-Aware":[85],"Text-Driven":[86],"(NATD-GSSL),":[89],"unified":[91],"framework":[92],"that":[93,108],"combines":[94],"construction,":[97],"refinement,":[99],"and":[100,142,155,228,242],"GSSL.":[101],"Our":[102,132],"follows":[104],"dual-graph":[106],"protocol":[107],"contrasts":[109],"noisy":[111,226],"derived":[113],"MedMentions":[115],"with":[116,186],"clean":[118,214],"Unified":[119],"Medical":[120],"Language":[121],"System":[122],"(UMLS)":[123],"reference":[124],"graph,":[125],"aligned":[126],"through":[127],"shared":[129],"gold":[130],"standard.":[131],"results":[133],"reveal":[134],"variability":[135],"across":[138],"both":[139],"pretext":[140],"tasks":[141],"Neural":[144],"Network":[145],"(GNN)":[146],"architectures.":[147],"Relation":[148],"reconstruction":[149,162],"is":[150,163],"highly":[151],"sensitive":[152],"to":[153,170,203,224,231],"benefits":[156],"well-defined":[158],"schemas,":[159],"whereas":[160],"feature":[161],"considerably":[164],"more":[165],"robust,":[166],"achieving":[167],"performance":[168],"comparable":[169],"clean-graph":[171],"settings.":[172],"Contrastive":[173],"objectives":[174],"are":[175,200,244],"generally":[176],"less":[177],"affected":[178],"by":[179],"but":[181],"depend":[182],"strongly":[183],"alignment":[185],"downstream":[187],"tasks.":[188],"GNN":[189],"architecture":[190],"also":[191],"plays":[192],"critical":[194],"role:":[195],"bidirectional":[196],"relational":[197,209],"message-passing":[198],"designs":[199],"better":[201],"suited":[202],"noisy,":[204],"graphs,":[206],"unidirectional":[208],"ones":[210],"perform":[211],"best":[212],"Overall,":[216],"NATD-GSSL":[217],"provides":[218],"practical":[219],"guidance":[220],"applying":[222],"real-world,":[225],"achieves":[229],"up":[230],"7\\%":[233],"improvement":[234],"over":[235],"pretrained":[236],"language":[237],"model":[238],"baselines.":[239],"All":[240],"code":[241],"benchmarks":[243],"publicly":[245],"available":[246],"at":[247],"https://github.com/OthmaneKabal/MC2GAE.":[248]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-05-09T00:00:00"}