{"id":"https://openalex.org/W4412888794","doi":"https://doi.org/10.18653/v1/2025.findings-acl.262","title":"BlockPruner: Fine-grained Pruning for Large Language Models","display_name":"BlockPruner: Fine-grained Pruning for Large Language Models","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888794","doi":"https://doi.org/10.18653/v1/2025.findings-acl.262"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.262","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.262","pdf_url":"https://aclanthology.org/2025.findings-acl.262.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.262.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5104251305","display_name":"Longguang Zhong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Longguang Zhong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101262140","display_name":"Fanqi Wan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fanqi Wan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091530162","display_name":"Ruijun Chen","orcid":"https://orcid.org/0000-0001-5281-4143"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ruijun Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040062188","display_name":"Xiaojun Quan","orcid":"https://orcid.org/0000-0002-8385-1083"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaojun Quan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5071266785","display_name":"Liangzhi Li","orcid":"https://orcid.org/0000-0002-8879-5957"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liangzhi Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":5.2763,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":{"value":0.95396322,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"5065","last_page":"5080"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9886999726295471,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9886999726295471,"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/T10028","display_name":"Topic Modeling","score":0.9771999716758728,"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":0.7170329689979553},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.6987694501876831},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4852711856365204},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4173603057861328},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4066387712955475}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7170329689979553},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.6987694501876831},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4852711856365204},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4173603057861328},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4066387712955475},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","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":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.262","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.262","pdf_url":"https://aclanthology.org/2025.findings-acl.262.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.262","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.262","pdf_url":"https://aclanthology.org/2025.findings-acl.262.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":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.6100000143051147}],"awards":[{"id":"https://openalex.org/G6792809764","display_name":null,"funder_award_id":"62176270","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G7447199816","display_name":null,"funder_award_id":"2023A1515012832","funder_id":"https://openalex.org/F4320337111","funder_display_name":"Basic and Applied Basic Research Foundation of Guangdong Province"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320337111","display_name":"Basic and Applied Basic Research Foundation of Guangdong Province","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888794.pdf","grobid_xml":"https://content.openalex.org/works/W4412888794.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/W3204019825"],"abstract_inverted_index":{"With":[0],"the":[1,5,14,42,59,63,72,121],"rapid":[2],"growth":[3],"in":[4,29,87],"size":[6],"and":[7,20,34,76,91,116,129,144,146,164],"complexity":[8],"of":[9,74,123,141,154],"large":[10],"language":[11],"models":[12],"(LLMs),":[13],"costs":[15],"associated":[16],"with":[17],"their":[18],"training":[19],"inference":[21],"have":[22,49],"escalated":[23],"significantly.Research":[24],"indicates":[25],"that":[26,78,159],"certain":[27],"layers":[28,37,64],"LLMs":[30,75,140],"harbor":[31],"substantial":[32],"redundancy,":[33],"pruning":[35,47,80,101,107,166],"these":[36,124],"has":[38],"minimal":[39],"impact":[40],"on":[41,53],"overall":[43],"performance.While":[44],"various":[45,142],"layer":[46,106,113],"methods":[48],"been":[50],"developed":[51],"based":[52],"this":[54,66],"insight,":[55],"they":[56],"generally":[57],"overlook":[58],"finer-grained":[60,79],"redundancies":[61,86],"within":[62],"themselves.In":[65],"paper,":[67],"we":[68],"delve":[69],"deeper":[70],"into":[71,114],"architecture":[73],"demonstrate":[77],"can":[81],"be":[82],"achieved":[83],"by":[84],"targeting":[85],"multi-head":[88],"attention":[89],"(MHA)":[90],"multi-layer":[92],"perceptron":[93],"(MLP)":[94],"blocks.We":[95],"propose":[96],"a":[97,131,151],"novel,":[98],"training-free":[99],"structured":[100],"approach":[102],"called":[103],"BlockPruner.Unlike":[104],"existing":[105],"methods,":[108],"BlockPruner":[109,138],"segments":[110],"each":[111],"Transformer":[112],"MHA":[115],"MLP":[117],"blocks.It":[118],"then":[119],"assesses":[120],"importance":[122],"blocks":[125],"using":[126],"perplexity":[127],"measures":[128],"applies":[130],"heuristic":[132],"search":[133],"for":[134],"iterative":[135],"pruning.We":[136],"applied":[137],"to":[139,168],"sizes":[143],"architectures":[145],"validated":[147],"its":[148],"performance":[149],"across":[150],"wide":[152],"range":[153],"downstream":[155],"tasks.Experimental":[156],"results":[157],"show":[158],"Block-Pruner":[160],"achieves":[161],"more":[162],"granular":[163],"effective":[165],"compared":[167],"state-of-the-art":[169],"baselines.":[170]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}