{"id":"https://openalex.org/W7154529990","doi":"https://doi.org/10.1007/s41060-026-01112-x","title":"RG-Hybrid: pure late fusion of RoBERTa and a graph transformer for robust, interpretable sentiment classification","display_name":"RG-Hybrid: pure late fusion of RoBERTa and a graph transformer for robust, interpretable sentiment classification","publication_year":2026,"publication_date":"2026-04-15","ids":{"openalex":"https://openalex.org/W7154529990","doi":"https://doi.org/10.1007/s41060-026-01112-x"},"language":"en","primary_location":{"id":"doi:10.1007/s41060-026-01112-x","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s41060-026-01112-x","pdf_url":"https://link.springer.com/content/pdf/10.1007/s41060-026-01112-x.pdf","source":{"id":"https://openalex.org/S4210195017","display_name":"International Journal of Data Science and Analytics","issn_l":"2364-415X","issn":["2364-415X","2364-4168"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319972","host_organization_name":"Springer International Publishing","host_organization_lineage":["https://openalex.org/P4310319972","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer International Publishing","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"International Journal of Data Science and Analytics","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://link.springer.com/content/pdf/10.1007/s41060-026-01112-x.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133712668","display_name":"Khaled Alahmadi","orcid":null},"institutions":[{"id":"https://openalex.org/I114017466","display_name":"University of Technology Sydney","ror":"https://ror.org/03f0f6041","country_code":"AU","type":"education","lineage":["https://openalex.org/I114017466"]},{"id":"https://openalex.org/I185163786","display_name":"King Abdulaziz University","ror":"https://ror.org/02ma4wv74","country_code":"SA","type":"education","lineage":["https://openalex.org/I185163786"]}],"countries":["AU","SA"],"is_corresponding":true,"raw_author_name":"Khaled Alahmadi","raw_affiliation_strings":["King Abdulaziz University, Jeddah, 21589, Saudi Arabia","University of Technology Sydney, Ultimo, NSW, 2007, Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"King Abdulaziz University, Jeddah, 21589, Saudi Arabia","institution_ids":["https://openalex.org/I185163786"]},{"raw_affiliation_string":"University of Technology Sydney, Ultimo, NSW, 2007, Australia","institution_ids":["https://openalex.org/I114017466"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133633768","display_name":"Sultan Alharbi","orcid":null},"institutions":[{"id":"https://openalex.org/I114017466","display_name":"University of Technology Sydney","ror":"https://ror.org/03f0f6041","country_code":"AU","type":"education","lineage":["https://openalex.org/I114017466"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Sultan Alharbi","raw_affiliation_strings":["University of Technology Sydney, Ultimo, NSW, 2007, Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Technology Sydney, Ultimo, NSW, 2007, Australia","institution_ids":["https://openalex.org/I114017466"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5133652027","display_name":"Xianzhi Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I114017466","display_name":"University of Technology Sydney","ror":"https://ror.org/03f0f6041","country_code":"AU","type":"education","lineage":["https://openalex.org/I114017466"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Xianzhi Wang","raw_affiliation_strings":["University of Technology Sydney, Ultimo, NSW, 2007, Australia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Technology Sydney, Ultimo, NSW, 2007, Australia","institution_ids":["https://openalex.org/I114017466"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5133712668"],"corresponding_institution_ids":["https://openalex.org/I114017466","https://openalex.org/I185163786"],"apc_list":{"value":2390,"currency":"EUR","value_usd":2990},"apc_paid":{"value":2390,"currency":"EUR","value_usd":2990},"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.43107594,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"22","issue":"1","first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.8032000064849854,"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/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.8032000064849854,"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":0.049300000071525574,"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/T12488","display_name":"Mental Health via Writing","score":0.02850000001490116,"subfield":{"id":"https://openalex.org/subfields/3207","display_name":"Social Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/negation","display_name":"Negation","score":0.5414999723434448},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.48570001125335693},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.48190000653266907},{"id":"https://openalex.org/keywords/overfitting","display_name":"Overfitting","score":0.4440000057220459},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.40619999170303345},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.40049999952316284},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.39559999108314514},{"id":"https://openalex.org/keywords/sentiment-analysis","display_name":"Sentiment analysis","score":0.36039999127388}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6710000038146973},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6306999921798706},{"id":"https://openalex.org/C2185349","wikidata":"https://www.wikidata.org/wiki/Q190558","display_name":"Negation","level":2,"score":0.5414999723434448},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.48570001125335693},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.48190000653266907},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4674000144004822},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.4440000057220459},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.40619999170303345},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.40049999952316284},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.39559999108314514},{"id":"https://openalex.org/C66402592","wikidata":"https://www.wikidata.org/wiki/Q2271421","display_name":"Sentiment analysis","level":2,"score":0.36039999127388},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34389999508857727},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.32989999651908875},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.3255999982357025},{"id":"https://openalex.org/C16311509","wikidata":"https://www.wikidata.org/wiki/Q4148050","display_name":"Dependency graph","level":3,"score":0.3025999963283539},{"id":"https://openalex.org/C197129107","wikidata":"https://www.wikidata.org/wiki/Q1921621","display_name":"Merge (version control)","level":2,"score":0.2996000051498413},{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.28790000081062317},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.2849999964237213},{"id":"https://openalex.org/C94124525","wikidata":"https://www.wikidata.org/wiki/Q912550","display_name":"Categorization","level":2,"score":0.2696000039577484},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.2662999927997589},{"id":"https://openalex.org/C17618745","wikidata":"https://www.wikidata.org/wiki/Q207509","display_name":"Copula (linguistics)","level":2,"score":0.2624000012874603},{"id":"https://openalex.org/C78023250","wikidata":"https://www.wikidata.org/wiki/Q657596","display_name":"Unary operation","level":2,"score":0.2583000063896179},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2538999915122986},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.2506999969482422}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1007/s41060-026-01112-x","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s41060-026-01112-x","pdf_url":"https://link.springer.com/content/pdf/10.1007/s41060-026-01112-x.pdf","source":{"id":"https://openalex.org/S4210195017","display_name":"International Journal of Data Science and Analytics","issn_l":"2364-415X","issn":["2364-415X","2364-4168"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319972","host_organization_name":"Springer International Publishing","host_organization_lineage":["https://openalex.org/P4310319972","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer International Publishing","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"International Journal of Data Science and Analytics","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1007/s41060-026-01112-x","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s41060-026-01112-x","pdf_url":"https://link.springer.com/content/pdf/10.1007/s41060-026-01112-x.pdf","source":{"id":"https://openalex.org/S4210195017","display_name":"International Journal of Data Science and Analytics","issn_l":"2364-415X","issn":["2364-415X","2364-4168"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319972","host_organization_name":"Springer International Publishing","host_organization_lineage":["https://openalex.org/P4310319972","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer International Publishing","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"International Journal of Data Science and Analytics","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.6967726945877075,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320334704","display_name":"Australian Research Council","ror":"https://ror.org/05mmh0f86"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7154529990.pdf","grobid_xml":"https://content.openalex.org/works/W7154529990.grobid-xml"},"referenced_works_count":33,"referenced_works":["https://openalex.org/W2160660844","https://openalex.org/W2163455955","https://openalex.org/W2788347302","https://openalex.org/W2962853356","https://openalex.org/W2963341956","https://openalex.org/W2963626623","https://openalex.org/W2964046515","https://openalex.org/W2970207740","https://openalex.org/W3035422918","https://openalex.org/W3035507081","https://openalex.org/W3152855017","https://openalex.org/W3156333129","https://openalex.org/W3167287584","https://openalex.org/W3173753074","https://openalex.org/W3176719207","https://openalex.org/W3201889554","https://openalex.org/W4206469492","https://openalex.org/W4213072767","https://openalex.org/W4225095674","https://openalex.org/W4229053887","https://openalex.org/W4286216949","https://openalex.org/W4304698333","https://openalex.org/W4312908198","https://openalex.org/W4317931697","https://openalex.org/W4384499195","https://openalex.org/W4389484816","https://openalex.org/W4392357044","https://openalex.org/W4400261301","https://openalex.org/W4401042286","https://openalex.org/W4402321055","https://openalex.org/W4403723818","https://openalex.org/W4406919742","https://openalex.org/W4408100028"],"related_works":[],"abstract_inverted_index":{"Abstract":[0],"Large":[1],"pretrained":[2],"language":[3],"models":[4],"(PLMs)":[5],"provide":[6],"strong":[7],"semantic":[8,50,119],"signals":[9],"for":[10],"sentiment":[11,139],"classification":[12],"but":[13,46,62],"process":[14],"text":[15],"as":[16,31],"linear":[17],"token":[18],"streams,":[19],"leaving":[20],"syntactic":[21],"relations":[22,44],"implicit.":[23],"This":[24,115],"limits":[25],"robustness":[26],"on":[27,74,137,197],"structure-sensitive":[28],"phenomena":[29],"such":[30,43],"negation":[32,188],"scope,":[33],"coordination,":[34],"and":[35,56,71,104,120,143,149,189],"clause-level":[36],"dependencies.":[37],"Graph":[38],"neural":[39],"networks":[40],"can":[41],"encode":[42],"explicitly":[45],"lack":[47],"the":[48,108,118,133],"broad":[49],"grounding":[51],"of":[52,128,163],"large-scale":[53],"pretraining.":[54],"Early-":[55],"mid-fusion":[57],"hybrids":[58],"address":[59],"this":[60],"gap":[61],"blend":[63],"modalities":[64],"through":[65,111],"learned":[66],"parameters,":[67],"obscuring":[68],"individual":[69],"contributions":[70,122],"risking":[72],"overfitting":[73],"smaller":[75],"datasets.":[76],"We":[77],"introduce":[78],"RG-Hybrid":[79,146],",":[80],"a":[81,86,91,154],"dual-branch":[82],"architecture":[83],"that":[84,179],"combines":[85],"fine-tuned":[87,155],"RoBERTa":[88],"encoder":[89],"with":[90,158],"graph":[92],"transformer":[93],"operating":[94,173],"over":[95,153],"dependency":[96,181],"parses.":[97],"The":[98],"two":[99],"branches":[100],"are":[101],"trained":[102],"separately":[103],"merged":[105],"only":[106],"at":[107],"decision":[109],"stage":[110],"parameter-free":[112],"logit":[113],"averaging.":[114],"design":[116],"keeps":[117],"structural":[121],"fully":[123,174],"separable,":[124],"enabling":[125],"direct":[126],"attribution":[127],"each":[129],"branch\u2019s":[130],"role":[131],"in":[132],"final":[134],"prediction.":[135],"Evaluated":[136],"four":[138],"benchmarks":[140],"spanning":[141],"document-level":[142],"sentence-level":[144],"corpora,":[145],"achieves":[147],"consistent":[148],"statistically":[150],"significant":[151],"improvements":[152],"RoBERTa-only":[156],"baseline,":[157],"relative":[159],"error":[160],"rate":[161],"reductions":[162],"up":[164],"to":[165],"17.66%.":[166],"It":[167],"outperforms":[168],"transductive":[169],"PLM\u2013GNN":[170],"baselines":[171],"despite":[172],"inductively.":[175],"Qualitative":[176],"analysis":[177],"confirms":[178],"explicit":[180],"structure":[182],"resolves":[183],"polarity":[184],"ambiguities":[185],"introduced":[186],"by":[187],"concessive":[190],"constructions,":[191],"while":[192],"also":[193],"identifying":[194],"failure":[195],"modes":[196],"ironic":[198],"text.":[199]},"counts_by_year":[],"updated_date":"2026-06-20T20:08:15.867695","created_date":"2026-04-16T00:00:00"}