{"id":"https://openalex.org/W4417105835","doi":"https://doi.org/10.48550/arxiv.2508.02532","title":"Contextual Graph Transformer: A Small Language Model for Enhanced Engineering Document Information Extraction","display_name":"Contextual Graph Transformer: A Small Language Model for Enhanced Engineering Document Information Extraction","publication_year":2025,"publication_date":"2025-08-04","ids":{"openalex":"https://openalex.org/W4417105835","doi":"https://doi.org/10.48550/arxiv.2508.02532"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2508.02532","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.02532","pdf_url":"https://arxiv.org/pdf/2508.02532","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2508.02532","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Reddy, Karan","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Reddy, Karan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5000550077","display_name":"Mayukha Pal","orcid":"https://orcid.org/0000-0001-6037-1338"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pal, Mayukha","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":2,"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":0.5230000019073486,"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":0.5230000019073486,"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":0.36090001463890076,"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":0.012799999676644802,"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/transformer","display_name":"Transformer","score":0.684499979019165},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.61080002784729},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4887999892234802},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.48030000925064087},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.42820000648498535},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.40779998898506165},{"id":"https://openalex.org/keywords/contextualization","display_name":"Contextualization","score":0.37860000133514404},{"id":"https://openalex.org/keywords/adaptability","display_name":"Adaptability","score":0.3467999994754791}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8034999966621399},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.684499979019165},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.61080002784729},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5656999945640564},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5622000098228455},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4887999892234802},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.48030000925064087},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.42820000648498535},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.40779998898506165},{"id":"https://openalex.org/C2780712339","wikidata":"https://www.wikidata.org/wiki/Q5165204","display_name":"Contextualization","level":3,"score":0.37860000133514404},{"id":"https://openalex.org/C177606310","wikidata":"https://www.wikidata.org/wiki/Q5674297","display_name":"Adaptability","level":2,"score":0.3467999994754791},{"id":"https://openalex.org/C195807954","wikidata":"https://www.wikidata.org/wiki/Q1662562","display_name":"Information extraction","level":2,"score":0.33809998631477356},{"id":"https://openalex.org/C145644426","wikidata":"https://www.wikidata.org/wiki/Q169411","display_name":"Unified Modeling Language","level":3,"score":0.31450000405311584},{"id":"https://openalex.org/C60048249","wikidata":"https://www.wikidata.org/wiki/Q37437","display_name":"Syntax","level":2,"score":0.3125},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.31200000643730164},{"id":"https://openalex.org/C71611378","wikidata":"https://www.wikidata.org/wiki/Q5165191","display_name":"Contextual design","level":3,"score":0.2987000048160553},{"id":"https://openalex.org/C130318100","wikidata":"https://www.wikidata.org/wiki/Q2268914","display_name":"Semantic similarity","level":2,"score":0.288100004196167},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.2705000042915344},{"id":"https://openalex.org/C2781235140","wikidata":"https://www.wikidata.org/wiki/Q275131","display_name":"Scratch","level":2,"score":0.2623000144958496},{"id":"https://openalex.org/C80156102","wikidata":"https://www.wikidata.org/wiki/Q788036","display_name":"Resolver","level":3,"score":0.25690001249313354},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.2540000081062317},{"id":"https://openalex.org/C2775955345","wikidata":"https://www.wikidata.org/wiki/Q7449071","display_name":"Semantic mapping","level":2,"score":0.2533000111579895},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.25060001015663147}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2508.02532","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.02532","pdf_url":"https://arxiv.org/pdf/2508.02532","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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"},{"id":"doi:10.48550/arxiv.2508.02532","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2508.02532","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2508.02532","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.02532","pdf_url":"https://arxiv.org/pdf/2508.02532","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Standard":[0],"transformer-based":[1],"language":[2,97],"models,":[3,91],"while":[4],"powerful":[5],"for":[6,45,70,110],"general":[7,166],"text,":[8],"often":[9,94],"struggle":[10],"with":[11,99,134],"the":[12,28],"fine-grained":[13],"syntax":[14],"and":[15,43,60,102,126,150,186],"entity":[16,184],"relationships":[17],"in":[18,189],"complex":[19],"technical,":[20],"engineering":[21],"documents.":[22],"To":[23],"address":[24],"this,":[25],"we":[26],"propose":[27],"Contextual":[29],"Graph":[30,39],"Transformer":[31,82],"(CGT),":[32],"a":[33,51,81,107,116,161],"hybrid":[34],"neural":[35],"architecture":[36],"that":[37],"combines":[38],"Neural":[40],"Networks":[41],"(GNNs)":[42],"Transformers":[44],"domain-specific":[46,172],"question":[47],"answering.":[48],"CGT":[49,105,121],"constructs":[50],"dynamic":[52],"graph":[53],"over":[54],"input":[55],"tokens":[56],"using":[57,160],"sequential,":[58],"skip-gram,":[59],"semantic":[61,152],"similarity":[62],"edges,":[63],"which":[64],"is":[65,156],"processed":[66],"by":[67,169],"GATv2Conv":[68],"layers":[69],"local":[71],"structure":[72,103],"learning.":[73],"These":[74],"enriched":[75],"embeddings":[76],"are":[77],"then":[78],"passed":[79],"to":[80,84,144,178],"encoder":[83],"capture":[85],"global":[86],"dependencies.":[87],"Unlike":[88],"generic":[89],"large":[90],"technical":[92,179],"domains":[93],"require":[95],"specialized":[96],"models":[98],"stronger":[100],"contextualization":[101],"awareness.":[104],"offers":[106],"parameter-efficient":[108],"solution":[109],"such":[111],"use":[112],"cases.":[113],"Integrated":[114],"into":[115],"Retrieval-Augmented":[117],"Generation":[118],"(RAG)":[119],"pipeline,":[120],"outperforms":[122],"baselines":[123],"like":[124],"GPT-2":[125,133],"BERT,":[127],"achieving":[128],"24.7%":[129],"higher":[130],"accuracy":[131],"than":[132],"62.4%":[135],"fewer":[136],"parameters.":[137],"This":[138,174],"gain":[139],"stems":[140],"from":[141,158],"CGTs":[142,176],"ability":[143],"jointly":[145],"model":[146,155],"structural":[147],"token":[148],"interactions":[149],"long-range":[151],"coherence.":[153],"The":[154],"trained":[157],"scratch":[159],"two-phase":[162],"approach:":[163],"pretraining":[164],"on":[165,171],"text":[167],"followed":[168],"fine-tuning":[170],"manuals.":[173],"highlights":[175],"adaptability":[177],"language,":[180],"enabling":[181],"better":[182],"grounding,":[183],"tracking,":[185],"retrieval-augmented":[187],"responses":[188],"real-world":[190],"applications.":[191]},"counts_by_year":[],"updated_date":"2026-04-21T08:09:41.155169","created_date":"2025-10-10T00:00:00"}