{"id":"https://openalex.org/W4381573231","doi":"https://doi.org/10.48550/arxiv.2306.11222","title":"LoSparse: Structured Compression of Large Language Models based on Low-Rank and Sparse Approximation","display_name":"LoSparse: Structured Compression of Large Language Models based on Low-Rank and Sparse Approximation","publication_year":2023,"publication_date":"2023-06-20","ids":{"openalex":"https://openalex.org/W4381573231","doi":"https://doi.org/10.48550/arxiv.2306.11222"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2306.11222","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.11222","pdf_url":"https://arxiv.org/pdf/2306.11222","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":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2306.11222","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5035808211","display_name":"Yixiao Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Li, Yixiao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090461985","display_name":"Yifan Yu","orcid":"https://orcid.org/0000-0002-8959-3169"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Yifan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059657111","display_name":"Qingru Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Qingru","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092230831","display_name":"Chen Liang","orcid":"https://orcid.org/0000-0002-1093-0362"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang, Chen","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101785116","display_name":"Pengcheng He","orcid":"https://orcid.org/0000-0003-3305-952X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Pengcheng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051745436","display_name":"Weizhu Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Weizhu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5109592386","display_name":"Tuo Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Tuo","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5035808211"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":7,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12303","display_name":"Tensor decomposition and applications","score":0.9825000166893005,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12303","display_name":"Tensor decomposition and applications","score":0.9825000166893005,"subfield":{"id":"https://openalex.org/subfields/2605","display_name":"Computational Mathematics"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9624000191688538,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.961899995803833,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.7506139278411865},{"id":"https://openalex.org/keywords/low-rank-approximation","display_name":"Low-rank approximation","score":0.7333284020423889},{"id":"https://openalex.org/keywords/rank","display_name":"Rank (graph theory)","score":0.7148312330245972},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.707680344581604},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4638490676879883},{"id":"https://openalex.org/keywords/sparse-matrix","display_name":"Sparse matrix","score":0.451399028301239},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4388800859451294},{"id":"https://openalex.org/keywords/matrix","display_name":"Matrix (chemical analysis)","score":0.42304107546806335},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3677724003791809},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.22916442155838013},{"id":"https://openalex.org/keywords/combinatorics","display_name":"Combinatorics","score":0.0673382580280304}],"concepts":[{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.7506139278411865},{"id":"https://openalex.org/C90199385","wikidata":"https://www.wikidata.org/wiki/Q6692777","display_name":"Low-rank approximation","level":3,"score":0.7333284020423889},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.7148312330245972},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.707680344581604},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4638490676879883},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.451399028301239},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4388800859451294},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.42304107546806335},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3677724003791809},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.22916442155838013},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0673382580280304},{"id":"https://openalex.org/C6557445","wikidata":"https://www.wikidata.org/wiki/Q173113","display_name":"Agronomy","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C159985019","wikidata":"https://www.wikidata.org/wiki/Q181790","display_name":"Composite material","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"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/C25023664","wikidata":"https://www.wikidata.org/wiki/Q1575637","display_name":"Hankel matrix","level":2,"score":0.0},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2306.11222","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.11222","pdf_url":"https://arxiv.org/pdf/2306.11222","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":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2306.11222","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2306.11222","pdf_url":null,"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":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2306.11222","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.11222","pdf_url":"https://arxiv.org/pdf/2306.11222","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":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"score":0.7300000190734863,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4381573231.pdf"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2373300491","https://openalex.org/W2378744544","https://openalex.org/W2594301978","https://openalex.org/W3144354057","https://openalex.org/W4385556248","https://openalex.org/W3213035342","https://openalex.org/W4301870758","https://openalex.org/W3213049004","https://openalex.org/W4294691692","https://openalex.org/W2789346341"],"abstract_inverted_index":{"Transformer":[0],"models":[1],"have":[2],"achieved":[3],"remarkable":[4],"results":[5],"in":[6,83,93],"various":[7],"natural":[8,118,124],"language":[9,119,125],"tasks,":[10],"but":[11],"they":[12],"are":[13],"often":[14],"prohibitively":[15],"large,":[16],"requiring":[17],"massive":[18],"memories":[19],"and":[20,27,36,56,69,80,90,102,123],"computational":[21],"resources.":[22],"To":[23],"reduce":[24],"the":[25,50,63,78,88,97],"size":[26],"complexity":[28],"of":[29,52,65,99],"these":[30],"models,":[31],"we":[32],"propose":[33],"LoSparse":[34],"(Low-Rank":[35],"Sparse":[37],"approximation),":[38],"a":[39,46,53,57],"novel":[40],"model":[41],"compression":[42,135],"technique":[43],"that":[44,130],"approximates":[45],"weight":[47],"matrix":[48,55],"by":[49],"sum":[51],"low-rank":[54,67,100,103],"sparse":[58],"matrix.":[59],"Our":[60],"method":[61,116],"combines":[62],"advantages":[64],"both":[66],"approximations":[68],"pruning,":[70],"while":[71,85],"avoiding":[72],"their":[73],"limitations.":[74],"Low-rank":[75],"approximation":[76,104],"compresses":[77],"coherent":[79],"expressive":[81,111],"parts":[82,92],"neurons,":[84],"pruning":[86,106],"removes":[87],"incoherent":[89],"non-expressive":[91],"neurons.":[94,112],"Pruning":[95],"enhances":[96],"diversity":[98],"approximations,":[101],"prevents":[105],"from":[107],"losing":[108],"too":[109],"many":[110],"We":[113,128],"evaluate":[114],"our":[115],"on":[117],"understanding,":[120],"question":[121],"answering,":[122],"generation":[126],"tasks.":[127],"show":[129],"it":[131],"significantly":[132],"outperforms":[133],"existing":[134],"methods.":[136]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":6}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2025-10-10T00:00:00"}