{"id":"https://openalex.org/W4412887764","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1035","title":"ShortGPT: Layers in Large Language Models are More Redundant Than You Expect","display_name":"ShortGPT: Layers in Large Language Models are More Redundant Than You Expect","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412887764","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1035"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.1035","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1035","pdf_url":"https://aclanthology.org/2025.findings-acl.1035.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.1035.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5109724260","display_name":"Xin Men","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xin Men","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018995861","display_name":"Mingyu Xu","orcid":"https://orcid.org/0009-0007-6347-0551"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mingyu Xu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087307843","display_name":"Qingyu Zhang","orcid":"https://orcid.org/0009-0009-4422-3971"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qingyu Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Qianhao Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qianhao Yuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067986384","display_name":"Bingning Wang","orcid":"https://orcid.org/0009-0007-7748-7098"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bingning Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101366403","display_name":"Hongyu Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hongyu Lin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100585978","display_name":"Yaojie Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yaojie Lu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100620300","display_name":"Xianpei Han","orcid":"https://orcid.org/0000-0002-1304-6302"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xianpei Han","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5115053594","display_name":"Weipeng Chen","orcid":"https://orcid.org/0000-0001-6396-9300"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Weipeng Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":9,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":43.9693,"has_fulltext":true,"cited_by_count":25,"citation_normalized_percentile":{"value":0.9979199,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"20192","last_page":"20204"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9553999900817871,"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":0.9553999900817871,"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":0.9075999855995178,"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":0.7101432681083679},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.3570888042449951}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7101432681083679},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.3570888042449951}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.1035","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1035","pdf_url":"https://aclanthology.org/2025.findings-acl.1035.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.1035","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1035","pdf_url":"https://aclanthology.org/2025.findings-acl.1035.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412887764.pdf","grobid_xml":"https://content.openalex.org/works/W4412887764.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"As":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs)":[4],"continue":[5],"to":[6,30,101,111,129],"advance,":[7],"their":[8,89],"computational":[9],"overhead":[10],"has":[11],"increased":[12],"significantly.In":[13],"this":[14,125],"study,":[15],"we":[16,37,67],"identify":[17],"notable":[18],"redundancy":[19,121],"across":[20,122],"the":[21,31,47,54,62],"layers":[22,27,86],"of":[23,49,64,120],"LLMs,":[24],"where":[25],"some":[26],"contribute":[28,128],"minimally":[29],"overall":[32],"network":[33],"functionality.To":[34],"quantify":[35],"this,":[36],"introduce":[38],"a":[39,117],"metric":[40],"called":[41],"Block":[42],"Influence":[43],"(BI),":[44],"which":[45],"measures":[46],"importance":[48],"each":[50],"layer":[51,65,107],"based":[52,87],"on":[53,61,88],"similarity":[55],"between":[56],"its":[57],"input":[58],"and":[59,79],"output.Based":[60],"observation":[63],"redundancy,":[66],"propose":[68],"straightforward":[69],"pruning":[70,98,114],"methods":[71,92],"for":[72,76,81,132],"different":[73],"tasks:":[74],"ShortGPT":[75],"multiple-choice":[77],"tasks":[78],"ShortGPT-gen":[80],"generative":[82],"tasks.They":[83],"prune":[84],"redundant":[85],"BI":[90],"scores.Our":[91],"demonstrate":[93],"superior":[94],"performance":[95],"over":[96],"previous":[97],"methods.The":[99],"ability":[100],"achieve":[102],"better":[103],"results":[104],"through":[105],"simple":[106],"pruning,":[108],"as":[109],"opposed":[110],"more":[112],"complex":[113],"techniques,":[115],"suggests":[116],"high":[118],"degree":[119],"layers.We":[123],"hope":[124],"work":[126],"will":[127],"future":[130],"research":[131],"improving":[133],"LLM":[134],"efficiency.":[135]},"counts_by_year":[{"year":2026,"cited_by_count":12},{"year":2025,"cited_by_count":13}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}