{"id":"https://openalex.org/W7105651591","doi":"https://doi.org/10.1109/ijcnn64981.2025.11228532","title":"Why Lift so Heavy? Slimming Large Language Models by Cutting Off the Layers","display_name":"Why Lift so Heavy? Slimming Large Language Models by Cutting Off the Layers","publication_year":2025,"publication_date":"2025-06-30","ids":{"openalex":"https://openalex.org/W7105651591","doi":"https://doi.org/10.1109/ijcnn64981.2025.11228532"},"language":null,"primary_location":{"id":"doi:10.1109/ijcnn64981.2025.11228532","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn64981.2025.11228532","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Joint Conference on Neural Networks (IJCNN)","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":null,"display_name":"Shuzhou Yuan","orcid":null},"institutions":[{"id":"https://openalex.org/I4401726909","display_name":"Center for Scalable Data Analytics and Artificial Intelligence","ror":"https://ror.org/01t4ttr56","country_code":"DE","type":"education","lineage":["https://openalex.org/I4401726909","https://openalex.org/I78650965","https://openalex.org/I926574661"]},{"id":"https://openalex.org/I78650965","display_name":"Technische Universit\u00e4t Dresden","ror":"https://ror.org/042aqky30","country_code":"DE","type":"education","lineage":["https://openalex.org/I78650965"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Shuzhou Yuan","raw_affiliation_strings":["ScaDS.AI and TU Dresden"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ScaDS.AI and TU Dresden","institution_ids":["https://openalex.org/I4401726909","https://openalex.org/I78650965"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Ercong Nie","orcid":null},"institutions":[{"id":"https://openalex.org/I3018771216","display_name":"LMU Klinikum","ror":"https://ror.org/02jet3w32","country_code":"DE","type":"healthcare","lineage":["https://openalex.org/I3018771216","https://openalex.org/I8204097"]},{"id":"https://openalex.org/I8204097","display_name":"Ludwig-Maximilians-Universit\u00e4t M\u00fcnchen","ror":"https://ror.org/05591te55","country_code":"DE","type":"education","lineage":["https://openalex.org/I8204097"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Ercong Nie","raw_affiliation_strings":["LMU Munich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"LMU Munich","institution_ids":["https://openalex.org/I3018771216","https://openalex.org/I8204097"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Bolei Ma","orcid":null},"institutions":[{"id":"https://openalex.org/I3018771216","display_name":"LMU Klinikum","ror":"https://ror.org/02jet3w32","country_code":"DE","type":"healthcare","lineage":["https://openalex.org/I3018771216","https://openalex.org/I8204097"]},{"id":"https://openalex.org/I8204097","display_name":"Ludwig-Maximilians-Universit\u00e4t M\u00fcnchen","ror":"https://ror.org/05591te55","country_code":"DE","type":"education","lineage":["https://openalex.org/I8204097"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Bolei Ma","raw_affiliation_strings":["LMU Munich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"LMU Munich","institution_ids":["https://openalex.org/I3018771216","https://openalex.org/I8204097"]}]},{"author_position":"last","author":{"id":null,"display_name":"Michael F\u00e4rber","orcid":null},"institutions":[{"id":"https://openalex.org/I4401726909","display_name":"Center for Scalable Data Analytics and Artificial Intelligence","ror":"https://ror.org/01t4ttr56","country_code":"DE","type":"education","lineage":["https://openalex.org/I4401726909","https://openalex.org/I78650965","https://openalex.org/I926574661"]},{"id":"https://openalex.org/I78650965","display_name":"Technische Universit\u00e4t Dresden","ror":"https://ror.org/042aqky30","country_code":"DE","type":"education","lineage":["https://openalex.org/I78650965"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Michael F\u00e4rber","raw_affiliation_strings":["ScaDS.AI and TU Dresden"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ScaDS.AI and TU Dresden","institution_ids":["https://openalex.org/I4401726909","https://openalex.org/I78650965"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"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":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.26820001006126404,"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":0.26820001006126404,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.1136000007390976,"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/T13629","display_name":"Text Readability and Simplification","score":0.08839999884366989,"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/lift","display_name":"Lift (data mining)","score":0.7021999955177307},{"id":"https://openalex.org/keywords/limiting","display_name":"Limiting","score":0.6463000178337097},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5687999725341797},{"id":"https://openalex.org/keywords/adaptability","display_name":"Adaptability","score":0.5361999869346619},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.4700999855995178},{"id":"https://openalex.org/keywords/work","display_name":"Work (physics)","score":0.45159998536109924}],"concepts":[{"id":"https://openalex.org/C139002025","wikidata":"https://www.wikidata.org/wiki/Q3001212","display_name":"Lift (data mining)","level":2,"score":0.7021999955177307},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.6463000178337097},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6004999876022339},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5687999725341797},{"id":"https://openalex.org/C177606310","wikidata":"https://www.wikidata.org/wiki/Q5674297","display_name":"Adaptability","level":2,"score":0.5361999869346619},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.4700999855995178},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.45159998536109924},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39800000190734863},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.38999998569488525},{"id":"https://openalex.org/C33347731","wikidata":"https://www.wikidata.org/wiki/Q285210","display_name":"Stacking","level":2,"score":0.3361999988555908},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.31130000948905945},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.31040000915527344},{"id":"https://openalex.org/C74672266","wikidata":"https://www.wikidata.org/wiki/Q815859","display_name":"Language acquisition","level":2,"score":0.2741999924182892},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.27059999108314514},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.2535000145435333}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/ijcnn64981.2025.11228532","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn64981.2025.11228532","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.8445069789886475}],"awards":[],"funders":[{"id":"https://openalex.org/F4320311649","display_name":"Ministry of Education","ror":"https://ror.org/036nq5137"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":18,"referenced_works":["https://openalex.org/W1970409510","https://openalex.org/W2070246124","https://openalex.org/W2955429306","https://openalex.org/W2964303116","https://openalex.org/W2970120757","https://openalex.org/W2979826702","https://openalex.org/W3136363192","https://openalex.org/W3153427360","https://openalex.org/W3168125510","https://openalex.org/W3172642864","https://openalex.org/W3173777717","https://openalex.org/W3174770825","https://openalex.org/W4205991051","https://openalex.org/W4220785642","https://openalex.org/W4226126941","https://openalex.org/W4385245566","https://openalex.org/W4385571003","https://openalex.org/W4402684117"],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"demonstrate":[4],"exceptional":[5],"language":[6],"understanding":[7],"and":[8,35,91],"generation":[9],"capabilities":[10],"by":[11],"learning":[12,19,82],"from":[13,49],"context.":[14],"Leveraging":[15],"the":[16,33,44,59,69,93,96,147],"strong":[17],"in-context":[18],"(ICL)":[20],"abilities":[21],"of":[22,37,46,61,71,78,150,164],"LLMs,":[23,38],"prompt-based":[24,76,116],"fine-tuning":[25,77,92,117],"has":[26],"proven":[27],"to":[28],"be":[29],"effective":[30],"for":[31,80,118,141,159],"enhancing":[32],"adaptability":[34],"alignment":[36],"especially":[39],"in":[40,52,75,115,123],"low-data":[41],"scenarios.":[42,83],"However,":[43],"billions":[45],"parameters":[47],"resulting":[48],"layer":[50,130],"stacking":[51],"LLMs":[53,79,107,151],"present":[54],"significant":[55],"computational":[56],"challenges,":[57],"limiting":[58],"practicality":[60],"fine-tuning.":[62],"To":[63],"tackle":[64],"this":[65],"problem,":[66],"we":[67,100],"explore":[68],"application":[70],"layer-wise":[72],"model":[73,89,94],"pruning":[74],"few-shot":[81],"Our":[84],"approach":[85],"involves":[86],"dropping":[87],"certain":[88,124],"layers":[90],"with":[95,104,127],"remaining":[97],"layers.":[98],"Surprisingly,":[99],"observe":[101],"that":[102],"even":[103],"fewer":[105],"layers,":[106],"maintain":[108],"similar":[109],"or":[110],"better":[111],"performance":[112],"levels,":[113],"particularly":[114],"text":[119],"classification":[120],"tasks.":[121],"Remarkably,":[122],"cases,":[125],"models":[126],"a":[128],"single":[129],"outperform":[131],"their":[132,154],"fully":[133],"layered":[134],"counterparts.":[135],"These":[136],"findings":[137],"offer":[138],"valuable":[139],"insights":[140],"future":[142],"work":[143],"aimed":[144],"at":[145],"mitigating":[146],"size":[148],"constraints":[149],"while":[152],"preserving":[153],"performance,":[155],"thereby":[156],"opening":[157],"avenues":[158],"significantly":[160],"more":[161],"efficient":[162],"use":[163],"LLMs.":[165]},"counts_by_year":[],"updated_date":"2026-07-22T07:51:19.307946","created_date":"2025-11-14T00:00:00"}