乐胖代购免代理版

{"id":"https://openalex.org/W4385571398","doi":"https://doi.org/10.18653/v1/2023.acl-long.95","title":"Fine-tuning Happens in Tiny Subspaces: Exploring Intrinsic Task-specific Subspaces of Pre-trained Language Models","display_name":"Fine-tuning Happens in Tiny Subspaces: Exploring Intrinsic Task-specific Subspaces of Pre-trained Language Models","publication_year":2023,"publication_date":"2023-01-01","ids":{"openalex":"https://openalex.org/W4385571398","doi":"https://doi.org/10.18653/v1/2023.acl-long.95"},"language":"en","primary_location":{"is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2023.acl-long.95","pdf_url":"https://aclanthology.org/2023.acl-long.95.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true},"type":"article","type_crossref":"proceedings-article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2023.acl-long.95.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100350965","display_name":"Zhong Zhang","orcid":"https://orcid.org/0000-0002-2993-8612"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]},{"id":"https://openalex.org/I4210145761","display_name":"Shenzhen Institutes of Advanced Technology","ror":"https://ror.org/04gh4er46","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210145761"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhong Zhang","raw_affiliation_strings":["Shenzhen Institute for Advanced Study, UESTC, Shenzhen, China","University of Electronic Science and Technology of China, Chengdu, China"],"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]},{"raw_affiliation_string":"Shenzhen Institute for Advanced Study, UESTC, Shenzhen, China","institution_ids":["https://openalex.org/I4210145761"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100691219","display_name":"Bang Liu","orcid":"https://orcid.org/0000-0002-2272-6852"},"institutions":[{"id":"https://openalex.org/I70931966","display_name":"Universit\u00e9 de Montr\u00e9al","ror":"https://ror.org/0161xgx34","country_code":"CA","type":"education","lineage":["https://openalex.org/I70931966"]},{"id":"https://openalex.org/I109736498","display_name":"Canadian Institute for Advanced Research","ror":"https://ror.org/01sdtdd95","country_code":"CA","type":"facility","lineage":["https://openalex.org/I109736498"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Bang Liu","raw_affiliation_strings":["Canada CIFAR AI Chair.","Mila & Universit\u00e9 de Montr\u00e9al, Montr\u00e9al, Canada"],"affiliations":[{"raw_affiliation_string":"Mila & Universit\u00e9 de Montr\u00e9al, Montr\u00e9al, Canada","institution_ids":["https://openalex.org/I70931966"]},{"raw_affiliation_string":"Canada CIFAR AI Chair.","institution_ids":["https://openalex.org/I109736498"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5088843448","display_name":"Junming Shao","orcid":"https://orcid.org/0000-0002-1183-2991"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]},{"id":"https://openalex.org/I4210145761","display_name":"Shenzhen Institutes of Advanced Technology","ror":"https://ror.org/04gh4er46","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210145761"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Junming Shao","raw_affiliation_strings":["Shenzhen Institute for Advanced Study, UESTC, Shenzhen, China","University of Electronic Science and Technology of China, Chengdu, China"],"affiliations":[{"raw_affiliation_string":"University of Electronic Science and Technology of China, Chengdu, China","institution_ids":["https://openalex.org/I150229711"]},{"raw_affiliation_string":"Shenzhen Institute for Advanced Study, UESTC, Shenzhen, China","institution_ids":["https://openalex.org/I4210145761"]}]}],"institution_assertions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.057,"has_fulltext":true,"fulltext_origin":"pdf","cited_by_count":3,"citation_normalized_percentile":{"value":0.999976,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":84,"max":88},"biblio":{"volume":null,"issue":null,"first_page":"1701","last_page":"1713"},"is_retracted":false,"is_paratext":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9942,"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/T10028","display_name":"Topic Modeling","score":0.9942,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9922,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.976,"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":[],"concepts":[{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.8295331},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.70934224},{"id":"https://openalex.org/C12362212","wikidata":"https://www.wikidata.org/wiki/Q728435","display_name":"Linear subspace","level":2,"score":0.7083862},{"id":"https://openalex.org/C165464430","wikidata":"https://www.wikidata.org/wiki/Q1570441","display_name":"Parameterized complexity","level":2,"score":0.66459215},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6509881},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.62638086},{"id":"https://openalex.org/C79337645","wikidata":"https://www.wikidata.org/wiki/Q779824","display_name":"Outlier","level":2,"score":0.5499378},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.51245743},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.50195765},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5010259},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.42352882},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.21418408},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.17016056},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2023.acl-long.95","pdf_url":"https://aclanthology.org/2023.acl-long.95.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true},{"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2305.17446","pdf_url":"https://arxiv.org/pdf/2305.17446","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":["Cornell University"],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false}],"best_oa_location":{"is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2023.acl-long.95","pdf_url":"https://aclanthology.org/2023.acl-long.95.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true},"sustainable_development_goals":[{"display_name":"Peace, justice, and strong institutions","score":0.71,"id":"https://metadata.un.org/sdg/16"}],"grants":[],"datasets":[],"versions":[],"referenced_works_count":36,"referenced_works":["https://openalex.org/W131533222","https://openalex.org/W2251939518","https://openalex.org/W2805003733","https://openalex.org/W2896457183","https://openalex.org/W2904243021","https://openalex.org/W2923014074","https://openalex.org/W2963477238","https://openalex.org/W2963748441","https://openalex.org/W2963846996","https://openalex.org/W2965373594","https://openalex.org/W2978670439","https://openalex.org/W2979826702","https://openalex.org/W3038012435","https://openalex.org/W3098042335","https://openalex.org/W3103754749","https://openalex.org/W3104162037","https://openalex.org/W3104263050","https://openalex.org/W3154229486","https://openalex.org/W3168867926","https://openalex.org/W3173902720","https://openalex.org/W3174770825","https://openalex.org/W3175752238","https://openalex.org/W3176647794","https://openalex.org/W3176828726","https://openalex.org/W3177323791","https://openalex.org/W3196813608","https://openalex.org/W3207663303","https://openalex.org/W3212072588","https://openalex.org/W4226070387","https://openalex.org/W4226110825","https://openalex.org/W4282939141","https://openalex.org/W4285242513","https://openalex.org/W4285272403","https://openalex.org/W4287122891","https://openalex.org/W4313908941","https://openalex.org/W4385574241"],"related_works":["https://openalex.org/W4287164812","https://openalex.org/W3213150849","https://openalex.org/W3172436493","https://openalex.org/W3100286349","https://openalex.org/W2957492749","https://openalex.org/W2896134808","https://openalex.org/W2386063599","https://openalex.org/W1995723671","https://openalex.org/W1975884855","https://openalex.org/W1887135636"],"abstract_inverted_index":{"Pre-trained":[0],"language":[1],"models":[2],"(PLMs)":[3],"are":[4,118],"known":[5],"to":[6,67,120,124],"be":[7,80],"overly":[8],"parameterized":[9],"and":[10,36],"have":[11],"significant":[12],"redundancy,":[13],"indicating":[14],"a":[15,40,58,87],"small":[16,88],"degree":[17],"of":[18,20,34,44,53,90],"freedom":[19],"the":[21,25,32,51,54,61,84,103,109],"PLMs.":[22],"Motivated":[23],"by":[24,49],"observation,":[26],"in":[27,83,102],"this":[28],"paper,":[29],"we":[30,94],"study":[31],"problem":[33],"re-parameterizing":[35],"fine-tuning":[37,55,101],"PLMs":[38,78],"from":[39],"new":[41],"perspective:":[42],"Discovery":[43],"intrinsic":[45,70],"task-specific":[46,71,122],"subspace.":[47,72,104],"Specifically,":[48],"exploiting":[50],"dynamics":[52],"process":[56],"for":[57],"given":[59],"task,":[60],"parameter":[62],"optimization":[63],"trajectory":[64],"is":[65,76],"learned":[66],"uncover":[68],"its":[69],"A":[73],"key":[74],"finding":[75],"that":[77,115],"can":[79],"effectively":[81],"fine-tuned":[82],"subspace":[85],"with":[86],"number":[89],"free":[91],"parameters.":[92],"Beyond,":[93],"observe":[95],"some":[96],"outlier":[97],"dimensions":[98,107,117],"emerging":[99],"during":[100],"Disabling":[105],"these":[106,116],"degrades":[108],"model":[110],"performance":[111],"significantly.":[112],"This":[113],"suggests":[114],"crucial":[119],"induce":[121],"knowledge":[123],"downstream":[125],"tasks.":[126]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W4385571398","counts_by_year":[{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":2}],"updated_date":"2025-01-02T12:21:19.019789","created_date":"2023-08-05"}