{"id":"https://openalex.org/W4415368114","doi":"https://doi.org/10.1109/isit63088.2025.11195417","title":"Deep Fusion: Capturing Dependencies in Contrastive Learning via Transformer Projection Heads","display_name":"Deep Fusion: Capturing Dependencies in Contrastive Learning via Transformer Projection Heads","publication_year":2025,"publication_date":"2025-06-22","ids":{"openalex":"https://openalex.org/W4415368114","doi":"https://doi.org/10.1109/isit63088.2025.11195417"},"language":null,"primary_location":{"id":"doi:10.1109/isit63088.2025.11195417","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isit63088.2025.11195417","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Symposium on Information Theory (ISIT)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5014205687","display_name":"Huanran Li","orcid":"https://orcid.org/0000-0002-9929-607X"},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]},{"id":"https://openalex.org/I2802999273","display_name":"Wisconsin Institutes for Discovery","ror":"https://ror.org/05gbven85","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I2802999273"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Huanran Li","raw_affiliation_strings":["Wisconsin Institute of Discovery, University of Wisconsin-Madison,Department of Electrical Engineering"],"affiliations":[{"raw_affiliation_string":"Wisconsin Institute of Discovery, University of Wisconsin-Madison,Department of Electrical Engineering","institution_ids":["https://openalex.org/I2802999273","https://openalex.org/I135310074"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5084976034","display_name":"Daniel Pimentel-Alarc\u00f3n","orcid":"https://orcid.org/0000-0003-0871-7009"},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]},{"id":"https://openalex.org/I2802999273","display_name":"Wisconsin Institutes for Discovery","ror":"https://ror.org/05gbven85","country_code":"US","type":"nonprofit","lineage":["https://openalex.org/I2802999273"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Daniel Pimentel-Alarc\u00f3n","raw_affiliation_strings":["Wisconsin Institute of Discovery, University of Wisconsin-Madison,Department of Biostatistics"],"affiliations":[{"raw_affiliation_string":"Wisconsin Institute of Discovery, University of Wisconsin-Madison,Department of Biostatistics","institution_ids":["https://openalex.org/I2802999273","https://openalex.org/I135310074"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5014205687"],"corresponding_institution_ids":["https://openalex.org/I135310074","https://openalex.org/I2802999273"],"apc_list":null,"apc_paid":null,"fwci":2.8331,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.92801174,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.8471999764442444,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.8471999764442444,"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/T14339","display_name":"Image Processing and 3D Reconstruction","score":0.8464000225067139,"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/T10320","display_name":"Neural Networks and Applications","score":0.8253999948501587,"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":0.6507999897003174},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.5590000152587891},{"id":"https://openalex.org/keywords/projection","display_name":"Projection (relational algebra)","score":0.5336999893188477},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.49000000953674316},{"id":"https://openalex.org/keywords/feature-extraction","display_name":"Feature extraction","score":0.48579999804496765},{"id":"https://openalex.org/keywords/feed-forward","display_name":"Feed forward","score":0.4027999937534332},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.3630000054836273},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.3578000068664551}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7347999811172485},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6571000218391418},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6507999897003174},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.5590000152587891},{"id":"https://openalex.org/C57493831","wikidata":"https://www.wikidata.org/wiki/Q3134666","display_name":"Projection (relational algebra)","level":2,"score":0.5336999893188477},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.49000000953674316},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.48579999804496765},{"id":"https://openalex.org/C38858127","wikidata":"https://www.wikidata.org/wiki/Q5441228","display_name":"Feed forward","level":2,"score":0.4027999937534332},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3630000054836273},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3578000068664551},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.32850000262260437},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.30720001459121704},{"id":"https://openalex.org/C65557600","wikidata":"https://www.wikidata.org/wiki/Q7249451","display_name":"Projection method","level":3,"score":0.30219998955726624},{"id":"https://openalex.org/C2780312720","wikidata":"https://www.wikidata.org/wiki/Q5689100","display_name":"Head (geology)","level":2,"score":0.2775000035762787},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.2759000062942505},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.27300000190734863},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2687000036239624},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2632000148296356},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.25920000672340393},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.2574999928474426},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.25200000405311584},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.250900000333786}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/isit63088.2025.11195417","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isit63088.2025.11195417","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Symposium on Information Theory (ISIT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":14,"referenced_works":["https://openalex.org/W1987971958","https://openalex.org/W2798991696","https://openalex.org/W3035524453","https://openalex.org/W3094502228","https://openalex.org/W3096609285","https://openalex.org/W3108655343","https://openalex.org/W3114632476","https://openalex.org/W3121523901","https://openalex.org/W3131500599","https://openalex.org/W3138516171","https://openalex.org/W3168822201","https://openalex.org/W4214481550","https://openalex.org/W4385245566","https://openalex.org/W4405102077"],"related_works":[],"abstract_inverted_index":{"Contrastive":[0],"Learning":[1],"(CL)":[2],"has":[3],"emerged":[4],"as":[5,41],"a":[6,22,39,73,116],"powerful":[7],"method":[8],"for":[9,55],"training":[10],"feature":[11],"extraction":[12],"models":[13],"using":[14,115],"unlabeled":[15],"data.":[16],"Recent":[17],"studies":[18],"suggest":[19],"that":[20,103],"incorporating":[21],"linear":[23],"projection":[24,43],"head":[25,44],"post-backbone":[26],"significantly":[27],"enhances":[28],"model":[29,105],"performance.":[30,64],"In":[31],"this":[32],"work,":[33],"we":[34,71,98],"investigate":[35],"the":[36,42,46,52,79,84,91,111],"use":[37],"of":[38,114],"Transformer":[40],"within":[45],"CL":[47],"framework,":[48],"aiming":[49],"to":[50,61,110],"exploit":[51],"Transformer's":[53],"capacity":[54],"capturing":[56],"long-range":[57],"dependencies":[58,87],"across":[59],"embeddings":[60],"further":[62],"improve":[63],"Through":[65],"both":[66],"experiments":[67],"and":[68],"theoretical":[69],"analysis,":[70],"reveal":[72],"compelling":[74],"\u201cDeep":[75],"Fusion\u201d":[76],"phenomenon":[77],"where":[78],"attention":[80],"mechanism":[81],"progressively":[82],"captures":[83],"correct":[85],"relational":[86],"among":[88],"samples":[89],"from":[90],"same":[92],"class":[93],"in":[94],"deeper":[95],"layers.":[96],"Finally,":[97],"demonstrate":[99],"through":[100],"experimental":[101],"results":[102],"our":[104],"achieves":[106],"superior":[107],"performance":[108],"compared":[109],"existing":[112],"approach":[113],"feedforward":[117],"layer.":[118]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-10-21T00:00:00"}