{"id":"https://openalex.org/W4411918560","doi":"https://doi.org/10.1186/s13636-025-00411-8","title":"Head information bottleneck (HIB): leveraging information bottleneck for efficient transformer head attribution and pruning","display_name":"Head information bottleneck (HIB): leveraging information bottleneck for efficient transformer head attribution and pruning","publication_year":2025,"publication_date":"2025-07-01","ids":{"openalex":"https://openalex.org/W4411918560","doi":"https://doi.org/10.1186/s13636-025-00411-8"},"language":"en","primary_location":{"id":"doi:10.1186/s13636-025-00411-8","is_oa":true,"landing_page_url":"https://doi.org/10.1186/s13636-025-00411-8","pdf_url":"https://asmp-eurasipjournals.springeropen.com/counter/pdf/10.1186/s13636-025-00411-8","source":{"id":"https://openalex.org/S19605986","display_name":"EURASIP Journal on Audio Speech and Music Processing","issn_l":"1687-4714","issn":["1687-4714","1687-4722"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310319965","host_organization_name":"Springer Nature","host_organization_lineage":["https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Nature"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"EURASIP Journal on Audio, Speech, and Music Processing","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","doaj"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://asmp-eurasipjournals.springeropen.com/counter/pdf/10.1186/s13636-025-00411-8","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5073118994","display_name":"Yukun Qian","orcid":"https://orcid.org/0000-0001-9095-4024"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yukun Qian","raw_affiliation_strings":["Key Laboratory for Key Technologies of IoT Terminals, Harbin Institute of Technology, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Key Laboratory for Key Technologies of IoT Terminals, Harbin Institute of Technology, Shenzhen, China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5044255594","display_name":"Xuyi Zhuang","orcid":null},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xuyi Zhuang","raw_affiliation_strings":["Key Laboratory for Key Technologies of IoT Terminals, Harbin Institute of Technology, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Key Laboratory for Key Technologies of IoT Terminals, Harbin Institute of Technology, Shenzhen, China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5009651653","display_name":"Mingjiang Wang","orcid":"https://orcid.org/0000-0002-4706-009X"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mingjiang Wang","raw_affiliation_strings":["Key Laboratory for Key Technologies of IoT Terminals, Harbin Institute of Technology, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Key Laboratory for Key Technologies of IoT Terminals, Harbin Institute of Technology, Shenzhen, China","institution_ids":["https://openalex.org/I204983213"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I204983213"],"apc_list":{"value":1115,"currency":"GBP","value_usd":1367},"apc_paid":{"value":1115,"currency":"GBP","value_usd":1367},"fwci":0.7655,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.71716747,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":"2025","issue":"1","first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9990000128746033,"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":0.9990000128746033,"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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.9901000261306763,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9896000027656555,"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/computer-science","display_name":"Computer science","score":0.8914180994033813},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.8605168461799622},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5956448316574097},{"id":"https://openalex.org/keywords/information-bottleneck-method","display_name":"Information bottleneck method","score":0.5760517716407776},{"id":"https://openalex.org/keywords/head","display_name":"Head (geology)","score":0.5486050248146057},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.4381585121154785},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4330137372016907},{"id":"https://openalex.org/keywords/mutual-information","display_name":"Mutual information","score":0.18830153346061707},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.07822656631469727},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.06745374202728271}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8914180994033813},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.8605168461799622},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5956448316574097},{"id":"https://openalex.org/C60008888","wikidata":"https://www.wikidata.org/wiki/Q6031013","display_name":"Information bottleneck method","level":3,"score":0.5760517716407776},{"id":"https://openalex.org/C2780312720","wikidata":"https://www.wikidata.org/wiki/Q5689100","display_name":"Head (geology)","level":2,"score":0.5486050248146057},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.4381585121154785},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4330137372016907},{"id":"https://openalex.org/C152139883","wikidata":"https://www.wikidata.org/wiki/Q252973","display_name":"Mutual information","level":2,"score":0.18830153346061707},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.07822656631469727},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.06745374202728271},{"id":"https://openalex.org/C114793014","wikidata":"https://www.wikidata.org/wiki/Q52109","display_name":"Geomorphology","level":1,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0},{"id":"https://openalex.org/C127313418","wikidata":"https://www.wikidata.org/wiki/Q1069","display_name":"Geology","level":0,"score":0.0},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1186/s13636-025-00411-8","is_oa":true,"landing_page_url":"https://doi.org/10.1186/s13636-025-00411-8","pdf_url":"https://asmp-eurasipjournals.springeropen.com/counter/pdf/10.1186/s13636-025-00411-8","source":{"id":"https://openalex.org/S19605986","display_name":"EURASIP Journal on Audio Speech and Music Processing","issn_l":"1687-4714","issn":["1687-4714","1687-4722"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310319965","host_organization_name":"Springer Nature","host_organization_lineage":["https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Nature"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"EURASIP Journal on Audio, Speech, and Music Processing","raw_type":"journal-article"},{"id":"pmh:oai:doaj.org/article:134775e6ecd2425aae40638c313e07a8","is_oa":true,"landing_page_url":"https://doaj.org/article/134775e6ecd2425aae40638c313e07a8","pdf_url":null,"source":{"id":"https://openalex.org/S4306401280","display_name":"DOAJ (DOAJ: Directory of Open Access Journals)","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":"repository"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"EURASIP Journal on Audio, Speech, and Music Processing, Vol 2025, Iss 1, Pp 1-13 (2025)","raw_type":"article"}],"best_oa_location":{"id":"doi:10.1186/s13636-025-00411-8","is_oa":true,"landing_page_url":"https://doi.org/10.1186/s13636-025-00411-8","pdf_url":"https://asmp-eurasipjournals.springeropen.com/counter/pdf/10.1186/s13636-025-00411-8","source":{"id":"https://openalex.org/S19605986","display_name":"EURASIP Journal on Audio Speech and Music Processing","issn_l":"1687-4714","issn":["1687-4714","1687-4722"],"is_oa":true,"is_in_doaj":true,"is_core":true,"host_organization":"https://openalex.org/P4310319965","host_organization_name":"Springer Nature","host_organization_lineage":["https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Nature"],"type":"journal"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"EURASIP Journal on Audio, Speech, and Music Processing","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1265899161","display_name":null,"funder_award_id":"62176102","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3929901318","display_name":null,"funder_award_id":"62276076","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4411918560.pdf","grobid_xml":"https://content.openalex.org/works/W4411918560.grobid-xml"},"referenced_works_count":27,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1849277567","https://openalex.org/W2726515241","https://openalex.org/W2962858109","https://openalex.org/W2964184826","https://openalex.org/W2972584841","https://openalex.org/W2996005046","https://openalex.org/W2996061341","https://openalex.org/W3038035611","https://openalex.org/W3093452197","https://openalex.org/W3096017728","https://openalex.org/W3097777922","https://openalex.org/W3133902514","https://openalex.org/W3173787059","https://openalex.org/W3186479087","https://openalex.org/W3197580070","https://openalex.org/W3208147779","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W4226380987","https://openalex.org/W4281492411","https://openalex.org/W4372347449","https://openalex.org/W4386142022","https://openalex.org/W4387966251","https://openalex.org/W4391235406","https://openalex.org/W6632575151","https://openalex.org/W6634564036"],"related_works":["https://openalex.org/W4401325445","https://openalex.org/W2622284819","https://openalex.org/W1504394672","https://openalex.org/W3089381707","https://openalex.org/W4285254085","https://openalex.org/W3034190530","https://openalex.org/W2741297526","https://openalex.org/W4295728955","https://openalex.org/W3129794609","https://openalex.org/W2304083841"],"abstract_inverted_index":{"Multi-head":[0],"attention":[1,39,43,69,79,102],"mechanisms":[2],"have":[3,19],"been":[4,21],"widely":[5],"applied":[6],"in":[7,15,28,41],"speech":[8],"pre-training.":[9],"However,":[10],"their":[11,115],"roles":[12],"and":[13,82,141,154],"effectiveness":[14,129],"various":[16],"downstream":[17,33],"tasks":[18],"not":[20],"fully":[22],"explored.":[23],"Attention":[24],"heads":[25,103,112],"may":[26],"vary":[27],"importance":[29],"depending":[30],"on":[31,89,114,161],"the":[32,38,42,48,54,60,64,72,83,99,105,110,127,136,142,152,159],"task.":[34,61],"We":[35,62,107],"assume":[36],"that":[37,56,134,147],"allocation":[40],"mechanism":[44],"is":[45],"similar":[46],"to":[47,52,70,87,97,104],"information":[49,65,76],"bottleneck,":[50],"aiming":[51],"highlight":[53],"parts":[55],"are":[57],"important":[58],"for":[59,120],"introduce":[63],"bottleneck":[66],"into":[67],"multi-head":[68],"estimate":[71],"degree":[73],"of":[74,101,130,135],"mutual":[75],"between":[77],"each":[78],"head\u2019s":[80],"output":[81],"input,":[84],"guiding":[85],"it":[86],"focus":[88],"useful":[90],"information.":[91],"Additionally,":[92],"we":[93],"propose":[94],"a":[95],"method":[96,132,140],"measure":[98],"contribution":[100],"tasks.":[106,163],"also":[108],"prune":[109],"model":[111,121],"based":[113],"contributions,":[116],"offering":[117],"interpretable":[118],"direction":[119],"pruning.":[122],"Our":[123],"experiments,":[124],"which":[125],"compared":[126],"pruning":[128],"our":[131,148],"with":[133,158],"traditional":[137],"Taylor":[138],"expansion":[139],"integrated":[143],"gradients":[144],"method,":[145],"show":[146],"approach":[149],"significantly":[150],"outperforms":[151],"former":[153],"achieves":[155],"comparable":[156],"results":[157],"latter":[160],"multiple":[162]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}