{"id":"https://openalex.org/W7164161587","doi":"https://doi.org/10.48550/arxiv.2606.10716","title":"Attention Expansion: Enhancing Keyphrase Extraction from Long Documents with Attention-Augmented Contextualized Embeddings","display_name":"Attention Expansion: Enhancing Keyphrase Extraction from Long Documents with Attention-Augmented Contextualized Embeddings","publication_year":2026,"publication_date":"2026-06-09","ids":{"openalex":"https://openalex.org/W7164161587","doi":"https://doi.org/10.48550/arxiv.2606.10716"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.10716","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.10716","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.2606.10716","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5070018836","display_name":"Roberto Mart\u00ednez-Cruz","orcid":"https://orcid.org/0009-0009-4264-1678"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mart\u00ednez-Cruz, Roberto","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063880854","display_name":"Alvaro J. L\u00f3pez-L\u00f3pez","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"L\u00f3pez-L\u00f3pez, Alvaro J.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138377088","display_name":"Jos\u00e9 Portela","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Portela, Jos\u00e9","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/T13083","display_name":"Advanced Text Analysis Techniques","score":0.9783999919891357,"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/T13083","display_name":"Advanced Text Analysis Techniques","score":0.9783999919891357,"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":0.0052999998442828655,"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/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.0032999999821186066,"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/salient","display_name":"Salient","score":0.6633999943733215},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.6121000051498413},{"id":"https://openalex.org/keywords/scope","display_name":"Scope (computer science)","score":0.6080999970436096},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6075999736785889},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.5756000280380249},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5681999921798706},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.5504999756813049},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.536300003528595},{"id":"https://openalex.org/keywords/mechanism","display_name":"Mechanism (biology)","score":0.44040000438690186}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8068000078201294},{"id":"https://openalex.org/C2780719617","wikidata":"https://www.wikidata.org/wiki/Q1030752","display_name":"Salient","level":2,"score":0.6633999943733215},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.6121000051498413},{"id":"https://openalex.org/C2778012447","wikidata":"https://www.wikidata.org/wiki/Q1034415","display_name":"Scope (computer science)","level":2,"score":0.6080999970436096},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6075999736785889},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5756000280380249},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5684000253677368},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5681999921798706},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.5504999756813049},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5478000044822693},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.536300003528595},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.44040000438690186},{"id":"https://openalex.org/C195807954","wikidata":"https://www.wikidata.org/wiki/Q1662562","display_name":"Information extraction","level":2,"score":0.36809998750686646},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.3458000123500824},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.329800009727478},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2863999903202057},{"id":"https://openalex.org/C155092808","wikidata":"https://www.wikidata.org/wiki/Q182557","display_name":"Computational linguistics","level":2,"score":0.2791000008583069},{"id":"https://openalex.org/C2778143727","wikidata":"https://www.wikidata.org/wiki/Q1820650","display_name":"Readability","level":2,"score":0.27619999647140503},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.275299996137619},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.26969999074935913},{"id":"https://openalex.org/C59656382","wikidata":"https://www.wikidata.org/wiki/Q191536","display_name":"Conjunction (astronomy)","level":2,"score":0.26510000228881836},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2583000063896179},{"id":"https://openalex.org/C2778751112","wikidata":"https://www.wikidata.org/wiki/Q835016","display_name":"Window (computing)","level":2,"score":0.2581999897956848},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.2549000084400177},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.2529999911785126}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.10716","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.10716","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.2606.10716","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.10716","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Pre-trained":[0],"language":[1,54],"models":[2,55,109,162],"(PLMs)":[3],"have":[4],"achieved":[5],"strong":[6],"performance":[7,155],"in":[8,167],"keyphrase":[9,29],"extraction":[10],"(KPE),":[11],"largely":[12],"due":[13],"to":[14,17,173],"their":[15,62,66],"ability":[16],"generate":[18],"rich":[19],"contextualized":[20],"representations.":[21],"However,":[22],"long-document":[23,208],"KPE":[24,108,154],"remains":[25],"challenging":[26],"because":[27],"salient":[28],"evidence":[30],"may":[31],"be":[32,40],"scattered":[33],"across":[34,122,156],"distant":[35],"document":[36],"sections":[37],"that":[38,83,149,181],"cannot":[39],"jointly":[41],"captured":[42],"within":[43],"the":[44,102,182],"limited":[45,193],"context":[46],"window":[47],"of":[48,106],"most":[49],"PLMs.":[50],"Although":[51],"long-context":[52,131,178],"large":[53],"(LLMs)":[56],"can":[57],"process":[58],"broader":[59],"textual":[60],"contexts,":[61],"computational":[63],"cost":[64],"limits":[65],"practicality":[67],"for":[68,192,207],"efficient":[69,203],"and":[70,130,137,143,163,176,204],"high-throughput":[71],"KPE.":[72,209],"To":[73],"overcome":[74],"this":[75],"limitation,":[76],"we":[77],"propose":[78],"an":[79,202],"attention":[80,113,150,199],"expansion":[81,151,200],"mechanism":[82,100,184],"augments":[84],"PLM":[85,124],"token":[86],"representations":[87],"with":[88],"information":[89,187],"from":[90,141],"surrounding":[91],"out-of-context":[92],"chunks":[93],"using":[94,133],"pre-trained":[95],"word":[96],"embeddings.":[97],"The":[98,170],"proposed":[99,183],"expands":[101],"effective":[103,205],"contextual":[104],"scope":[105],"PLM-based":[107],"without":[110],"requiring":[111],"full-document":[112],"or":[114],"expensive":[115],"LLM-based":[116],"inference.":[117],"We":[118],"evaluate":[119],"our":[120],"approach":[121],"five":[123,138],"backbones,":[125],"including":[126],"general-purpose,":[127],"scientific,":[128],"task-specific,":[129],"encoders,":[132],"two":[134],"training":[135],"regimes":[136],"benchmark":[139],"corpora":[140],"scientific":[142],"news":[144],"domains.":[145],"Experimental":[146],"results":[147,197],"demonstrate":[148],"consistently":[152],"enhances":[153],"all":[157],"evaluation":[158],"settings,":[159],"outperforming":[160],"state-of-the-art":[161],"yielding":[164],"notable":[165],"improvements":[166,171],"F1":[168],"score.":[169],"extend":[172],"domain-specific,":[174],"task-specialized,":[175],"native":[177],"models,":[179],"showing":[180],"provides":[185],"complementary":[186],"rather":[188],"than":[189],"merely":[190],"compensating":[191],"input":[194],"length.":[195],"These":[196],"establish":[198],"as":[201],"strategy":[206]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-11T00:00:00"}