{"id":"https://openalex.org/W4308234147","doi":"https://doi.org/10.1109/icip46576.2022.9897905","title":"Rethinking Efficacy of Softmax for Lightweight Non-local Neural Networks","display_name":"Rethinking Efficacy of Softmax for Lightweight Non-local Neural Networks","publication_year":2022,"publication_date":"2022-10-16","ids":{"openalex":"https://openalex.org/W4308234147","doi":"https://doi.org/10.1109/icip46576.2022.9897905"},"language":"en","primary_location":{"id":"doi:10.1109/icip46576.2022.9897905","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icip46576.2022.9897905","pdf_url":null,"source":{"id":"https://openalex.org/S4363607719","display_name":"2022 IEEE International Conference on Image Processing (ICIP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE International Conference on Image Processing (ICIP)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5033296849","display_name":"Yooshin Cho","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yooshin Cho","raw_affiliation_strings":["KAIST,School of Electrical Engineering,South Korea","School of Electrical Engineering, KAIST, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST,School of Electrical Engineering,South Korea","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"School of Electrical Engineering, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100395865","display_name":"Youngsoo Kim","orcid":"https://orcid.org/0000-0002-2828-8398"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Youngsoo Kim","raw_affiliation_strings":["KAIST,School of Electrical Engineering,South Korea","School of Electrical Engineering, KAIST, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST,School of Electrical Engineering,South Korea","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"School of Electrical Engineering, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053657112","display_name":"Hanbyel Cho","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hanbyel Cho","raw_affiliation_strings":["KAIST,School of Electrical Engineering,South Korea","School of Electrical Engineering, KAIST, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST,School of Electrical Engineering,South Korea","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"School of Electrical Engineering, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081141535","display_name":"Jaesung Ahn","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I3130020584","display_name":"International Graduate School of English","ror":"https://ror.org/0181xpd52","country_code":"KR","type":"education","lineage":["https://openalex.org/I3130020584"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jaesung Ahn","raw_affiliation_strings":["KAIST,Kim Jaechul Graduate School of AI,South Korea","Kim Jaechul Graduate School of AI, KAIST, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST,Kim Jaechul Graduate School of AI,South Korea","institution_ids":["https://openalex.org/I157485424","https://openalex.org/I3130020584"]},{"raw_affiliation_string":"Kim Jaechul Graduate School of AI, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424","https://openalex.org/I3130020584"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011187138","display_name":"Hyeong Gwon Hong","orcid":"https://orcid.org/0000-0003-2410-2717"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I3130020584","display_name":"International Graduate School of English","ror":"https://ror.org/0181xpd52","country_code":"KR","type":"education","lineage":["https://openalex.org/I3130020584"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyeong Gwon Hong","raw_affiliation_strings":["KAIST,Kim Jaechul Graduate School of AI,South Korea","Kim Jaechul Graduate School of AI, KAIST, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST,Kim Jaechul Graduate School of AI,South Korea","institution_ids":["https://openalex.org/I157485424","https://openalex.org/I3130020584"]},{"raw_affiliation_string":"Kim Jaechul Graduate School of AI, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424","https://openalex.org/I3130020584"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100606266","display_name":"Junmo Kim","orcid":"https://orcid.org/0000-0002-7174-7932"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I3130020584","display_name":"International Graduate School of English","ror":"https://ror.org/0181xpd52","country_code":"KR","type":"education","lineage":["https://openalex.org/I3130020584"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Junmo Kim","raw_affiliation_strings":["KAIST,School of Electrical Engineering,South Korea","Kim Jaechul Graduate School of AI, KAIST, South Korea","School of Electrical Engineering, KAIST, South Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST,School of Electrical Engineering,South Korea","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"Kim Jaechul Graduate School of AI, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424","https://openalex.org/I3130020584"]},{"raw_affiliation_string":"School of Electrical Engineering, KAIST, South Korea","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.2774,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.61274386,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"1031","last_page":"1035"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9973999857902527,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9959999918937683,"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/softmax-function","display_name":"Softmax function","score":0.9363216757774353},{"id":"https://openalex.org/keywords/initialization","display_name":"Initialization","score":0.7430449724197388},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.7313020825386047},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.7140583992004395},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.6802719831466675},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6719784140586853},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.5489363074302673},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5276647210121155},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5067818760871887},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.46649810671806335},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.44436508417129517},{"id":"https://openalex.org/keywords/feature-vector","display_name":"Feature vector","score":0.43888619542121887},{"id":"https://openalex.org/keywords/block-size","display_name":"Block size","score":0.4352377653121948},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.3587538003921509},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.2854764461517334},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.19545620679855347}],"concepts":[{"id":"https://openalex.org/C188441871","wikidata":"https://www.wikidata.org/wiki/Q7554146","display_name":"Softmax function","level":3,"score":0.9363216757774353},{"id":"https://openalex.org/C114466953","wikidata":"https://www.wikidata.org/wiki/Q6034165","display_name":"Initialization","level":2,"score":0.7430449724197388},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.7313020825386047},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.7140583992004395},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6802719831466675},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6719784140586853},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.5489363074302673},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5276647210121155},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5067818760871887},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.46649810671806335},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.44436508417129517},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.43888619542121887},{"id":"https://openalex.org/C41431624","wikidata":"https://www.wikidata.org/wiki/Q1053357","display_name":"Block size","level":3,"score":0.4352377653121948},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3587538003921509},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2854764461517334},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.19545620679855347},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icip46576.2022.9897905","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icip46576.2022.9897905","pdf_url":null,"source":{"id":"https://openalex.org/S4363607719","display_name":"2022 IEEE International Conference on Image Processing (ICIP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2022 IEEE International Conference on Image Processing (ICIP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":29,"referenced_works":["https://openalex.org/W1677182931","https://openalex.org/W2302255633","https://openalex.org/W2551176409","https://openalex.org/W2896457183","https://openalex.org/W2963091558","https://openalex.org/W2963918968","https://openalex.org/W2964110616","https://openalex.org/W2964137095","https://openalex.org/W2981689412","https://openalex.org/W2996428491","https://openalex.org/W3026674654","https://openalex.org/W3030728803","https://openalex.org/W3033529678","https://openalex.org/W3035050085","https://openalex.org/W3094502228","https://openalex.org/W3118608800","https://openalex.org/W3179869055","https://openalex.org/W3210646981","https://openalex.org/W4289238196","https://openalex.org/W4385245566","https://openalex.org/W6739901393","https://openalex.org/W6755207826","https://openalex.org/W6756712531","https://openalex.org/W6768021236","https://openalex.org/W6777804712","https://openalex.org/W6779163297","https://openalex.org/W6784333009","https://openalex.org/W6787972765","https://openalex.org/W6803508786"],"related_works":["https://openalex.org/W3107204728","https://openalex.org/W4287591324","https://openalex.org/W4226420367","https://openalex.org/W2980176872","https://openalex.org/W2962876041","https://openalex.org/W3090555870","https://openalex.org/W3108503355","https://openalex.org/W2249953602","https://openalex.org/W2801655600","https://openalex.org/W2912971006"],"abstract_inverted_index":{"Non-local":[0],"(NL)":[1],"block":[2,17,32],"is":[3,27,74,101,107],"a":[4],"popular":[5],"module":[6],"that":[7,73],"demonstrates":[8],"the":[9,31,42,51,61,68,79,83,104,114],"capability":[10],"to":[11,29,33,40,77,132],"model":[12],"global":[13],"contexts.":[14],"However,":[15],"NL":[16,45,84],"generally":[18,75],"has":[19],"heavy":[20],"computation":[21],"and":[22,53,99,124,136],"memory":[23],"costs,":[24],"so":[25],"it":[26],"impractical":[28],"apply":[30],"high-resolution":[34],"feature":[35,57],"maps.":[36],"In":[37,126],"this":[38],"paper,":[39],"investigate":[41],"efficacy":[43],"of":[44,55,70,82,96],"block,":[46],"we":[47,117],"empirically":[48],"analyze":[49],"if":[50,103],"magnitude":[52,95,105],"direction":[54],"input":[56],"vectors":[58],"properly":[59],"affect":[60],"attention":[62,80,145],"between":[63],"vectors.":[64],"The":[65],"results":[66],"show":[67],"inefficacy":[69],"softmax":[71,90,111],"operation":[72,91,112],"used":[76],"normalize":[78],"map":[81],"block.":[85],"Attention":[86],"maps":[87],"normalized":[88],"with":[89,113],"highly":[92],"rely":[93],"upon":[94],"key":[97],"vectors,":[98],"performance":[100,120],"degenerated":[102],"information":[106],"removed.":[108],"By":[109],"replacing":[110],"scaling":[115],"factor,":[116],"demonstrate":[118],"improved":[119],"on":[121],"CIFAR-10,":[122],"CIFAR-100,":[123],"Tiny-ImageNet.":[125],"Addition,":[127],"our":[128,141],"method":[129,142],"shows":[130],"robustness":[131],"embedding":[133,137],"channel":[134],"reduction":[135],"weight":[138],"initialization.":[139],"Notably,":[140],"makes":[143],"multi-head":[144],"employable":[146],"without":[147],"additional":[148],"computational":[149],"cost.":[150]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}