{"id":"https://openalex.org/W4408151357","doi":"https://doi.org/10.1145/3658617.3697648","title":"Learning to Prune and Low-Rank Adaptation for Compact Language Model Deployment","display_name":"Learning to Prune and Low-Rank Adaptation for Compact Language Model Deployment","publication_year":2025,"publication_date":"2025-01-20","ids":{"openalex":"https://openalex.org/W4408151357","doi":"https://doi.org/10.1145/3658617.3697648"},"language":"en","primary_location":{"id":"doi:10.1145/3658617.3697648","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3658617.3697648","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th Asia and South Pacific Design Automation Conference","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":"https://openalex.org/A5036983020","display_name":"Asmer Hamid Ali","orcid":null},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Asmer Hamid Ali","raw_affiliation_strings":["School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona, USA"],"raw_orcid":"https://orcid.org/0009-0003-5291-0487","affiliations":[{"raw_affiliation_string":"School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona, USA","institution_ids":["https://openalex.org/I55732556"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074715434","display_name":"Fan Zhang","orcid":"https://orcid.org/0000-0002-6823-2700"},"institutions":[{"id":"https://openalex.org/I145311948","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95","country_code":"US","type":"education","lineage":["https://openalex.org/I145311948"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Fan Zhang","raw_affiliation_strings":["Johns Hopkins Univ., Baltimore, MD, USA"],"raw_orcid":"https://orcid.org/0000-0002-6823-2700","affiliations":[{"raw_affiliation_string":"Johns Hopkins Univ., Baltimore, MD, USA","institution_ids":["https://openalex.org/I145311948"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100419157","display_name":"Li Yang","orcid":"https://orcid.org/0000-0002-2839-6196"},"institutions":[{"id":"https://openalex.org/I102149020","display_name":"University of North Carolina at Charlotte","ror":"https://ror.org/04dawnj30","country_code":"US","type":"education","lineage":["https://openalex.org/I102149020"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Li Yang","raw_affiliation_strings":["University of North Carolina, Charlotte, North Carolina, USA"],"raw_orcid":"https://orcid.org/0000-0002-2839-6196","affiliations":[{"raw_affiliation_string":"University of North Carolina, Charlotte, North Carolina, USA","institution_ids":["https://openalex.org/I102149020"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5047916979","display_name":"Deliang Fan","orcid":"https://orcid.org/0000-0002-7989-6297"},"institutions":[{"id":"https://openalex.org/I55732556","display_name":"Arizona State University","ror":"https://ror.org/03efmqc40","country_code":"US","type":"education","lineage":["https://openalex.org/I55732556"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Deliang Fan","raw_affiliation_strings":["School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona, USA"],"raw_orcid":"https://orcid.org/0000-0002-7989-6297","affiliations":[{"raw_affiliation_string":"School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona, USA","institution_ids":["https://openalex.org/I55732556"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"36","last_page":"42"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9970999956130981,"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":0.9970999956130981,"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":0.9799000024795532,"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":0.977400004863739,"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/software-deployment","display_name":"Software deployment","score":0.7542926073074341},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.716829776763916},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.7083213329315186},{"id":"https://openalex.org/keywords/rank","display_name":"Rank (graph theory)","score":0.6069512963294983},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.34313899278640747},{"id":"https://openalex.org/keywords/software-engineering","display_name":"Software engineering","score":0.1805112063884735},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.08021354675292969},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.07970565557479858}],"concepts":[{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.7542926073074341},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.716829776763916},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.7083213329315186},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.6069512963294983},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.34313899278640747},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.1805112063884735},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.08021354675292969},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.07970565557479858},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.0},{"id":"https://openalex.org/C169760540","wikidata":"https://www.wikidata.org/wiki/Q207011","display_name":"Neuroscience","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3658617.3697648","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3658617.3697648","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 30th Asia and South Pacific Design Automation Conference","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2811135852","display_name":"FET: Small: AlignMEM: Fast and Efficient DNA Sequence Alignment in Non-Volatile Magnetic RAM","funder_award_id":"2349802","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G4598971362","display_name":"Collaborative Research: FuSe: Efficient Situation-Aware AI Processing in Advanced 2-Terminal SOT-MRAM","funder_award_id":"2414603","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G5407660188","display_name":"ACED Fab: Ultrafast, low-power AI chip with a new class of MRAM for learning and inference at edge","funder_award_id":"2314591","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7748010564","display_name":"CAREER: Efficient, Dynamic, Robust, and On-Device Continual Deep Learning with Non-Volatile Memory based In-Memory Computing System","funder_award_id":"2342726","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":64,"referenced_works":["https://openalex.org/W1994530392","https://openalex.org/W2119144962","https://openalex.org/W2618305643","https://openalex.org/W2750779823","https://openalex.org/W2799054028","https://openalex.org/W2890894339","https://openalex.org/W2896457183","https://openalex.org/W2923014074","https://openalex.org/W2946609015","https://openalex.org/W2950151997","https://openalex.org/W2953271402","https://openalex.org/W2965373594","https://openalex.org/W2978017171","https://openalex.org/W2998617917","https://openalex.org/W3027440908","https://openalex.org/W3042900599","https://openalex.org/W3044749682","https://openalex.org/W3104186312","https://openalex.org/W3104263050","https://openalex.org/W3121696187","https://openalex.org/W3168867926","https://openalex.org/W3205949070","https://openalex.org/W3209304705","https://openalex.org/W4221163429","https://openalex.org/W4221164335","https://openalex.org/W4225401825","https://openalex.org/W4283811336","https://openalex.org/W4287715551","https://openalex.org/W4292779060","https://openalex.org/W4297813615","https://openalex.org/W4306705029","https://openalex.org/W4308288764","https://openalex.org/W4313484599","https://openalex.org/W4321276845","https://openalex.org/W4322718191","https://openalex.org/W4327672335","https://openalex.org/W4361230739","https://openalex.org/W4364387438","https://openalex.org/W4377371819","https://openalex.org/W4378509449","https://openalex.org/W4378942328","https://openalex.org/W4380715573","https://openalex.org/W4381586827","https://openalex.org/W4384389802","https://openalex.org/W4385565413","https://openalex.org/W4386566659","https://openalex.org/W4386978016","https://openalex.org/W4387561365","https://openalex.org/W4388183436","https://openalex.org/W4390041933","https://openalex.org/W4390306161","https://openalex.org/W4390874728","https://openalex.org/W4391799377","https://openalex.org/W4391871574","https://openalex.org/W6677103964","https://openalex.org/W6729448088","https://openalex.org/W6778883912","https://openalex.org/W6788034315","https://openalex.org/W6796581206","https://openalex.org/W6852962002","https://openalex.org/W6853251322","https://openalex.org/W6854475153","https://openalex.org/W6860041859","https://openalex.org/W6861489518"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2770234245","https://openalex.org/W96612179","https://openalex.org/W4229499248","https://openalex.org/W2566006169","https://openalex.org/W1567818861","https://openalex.org/W2987774938","https://openalex.org/W4256492088"],"abstract_inverted_index":{"Nowadays,":[0],"parameter-efficient":[1],"fine-tuning":[2,40,80],"(PEFT)":[3],"large":[4,49],"pre-trained":[5],"models":[6,225],"(LPMs)":[7],"for":[8,33,46,79,100,147],"downstream":[9,36,81,101],"task":[10],"have":[11],"gained":[12],"significant":[13],"popularity,":[14],"since":[15],"it":[16],"could":[17,128],"significantly":[18],"minimize":[19],"the":[20,41,48,52,70,97,107,122,130,137,155,171,216],"training":[21,138],"computational":[22],"overhead.":[23],"The":[24,189,202],"representative":[25],"work,":[26,67],"LoRA":[27],"[1],":[28],"learns":[29],"a":[30,34,74,112],"low-rank":[31,87,91,98,133],"adaptor":[32],"new":[35],"task,":[37],"rather":[38],"than":[39],"whole":[42],"backbone":[43,123,144],"model.":[44],"However,":[45],"inference,":[47],"size":[50,140,146],"of":[51,114,132,168,179,194,218],"learned":[53],"model":[54,157,204],"remains":[55],"unchanged,":[56],"leading":[57],"to":[58,72,110,135,207],"in-efficient":[59],"inference":[60,149,181,196,209],"computation.":[61,150],"To":[62],"mitigate":[63],"this,":[64],"in":[65,184],"this":[66],"we":[68],"are":[69],"first":[71],"propose":[73],"learning-to-prune":[75],"methodology":[76],"specially":[77],"designed":[78],"tasks":[82],"based":[83],"on":[84],"LPMs":[85],"with":[86,158,176,199,212],"adaptation.":[88],"Unlike":[89],"prior":[90],"adaptation":[92,134],"approaches":[93],"that":[94,119,154],"only":[95],"learn":[96,111],"adaptors":[99],"tasks,":[102],"our":[103,126,159,219],"method":[104,127,160],"further":[105],"leverages":[106],"Gumbel-Sigmoid":[108],"tricks":[109],"set":[113],"trainable":[115],"binary":[116],"channel-wise":[117,164],"masks":[118],"automatically":[120],"prune":[121],"LPMs.":[124],"Therefore,":[125],"leverage":[129],"benefits":[131],"reduce":[136],"parameters":[139],"and":[141,226],"smaller":[142],"pruned":[143],"LPM":[145],"efficient":[148],"Extensive":[151],"experiments":[152],"show":[153],"Pruned-RoBbase":[156],"achieves":[161,205],"an":[162,177,192],"average":[163,178,193],"structured":[165],"pruning":[166,221],"ratio":[167],"24.5%":[169,213],"across":[170,223],"popular":[172],"GLUE":[173],"Benchmark,":[174],"coupled":[175],"18%":[180],"time":[182,197,210],"speed-up":[183],"real":[185],"NVIDIA":[186],"A5000":[187],"GPU.":[188],"Pruned-DistilBERT":[190],"shows":[191],"13%":[195],"improvement":[198,211],"17%":[200],"sparsity.":[201],"Pruned-LLaMA-7B":[203],"up":[206],"18.2%":[208],"sparsity,":[214],"demonstrating":[215],"effectiveness":[217],"learnable":[220],"approach":[222],"different":[224],"tasks.":[227]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}