{"id":"https://openalex.org/W7217744112","doi":null,"title":"LLaMA-2 and LLaMA-3 Models with SparseGPT-Based Model Compression and LoRA/QLoRA Integration: Efficient Large Language Model Optimization for Resource-Constrained Environments","display_name":"LLaMA-2 and LLaMA-3 Models with SparseGPT-Based Model Compression and LoRA/QLoRA Integration: Efficient Large Language Model Optimization for Resource-Constrained Environments","publication_year":2025,"publication_date":"2025-09-08","ids":{"openalex":"https://openalex.org/W7217744112"},"language":"en","primary_location":{"id":"pmh:oai:oai.as-proceeding.com:article/2810","doi":null,"is_oa":true,"landing_page_url":"https://as-proceeding.com/index.php/ijanser/article/view/2810","pdf_url":"https://as-proceeding.com/index.php/ijanser/article/download/2810/2677","source":{"id":"https://openalex.org/S4387288395","display_name":"International Journal of Advanced Natural Sciences and Engineering Researches","issn_l":"2980-0811","issn":["2980-0811"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2980-0811","raw_type":"info:eu-repo/semantics/article"},"volume":null,"type":"article","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://as-proceeding.com/index.php/ijanser/article/download/2810/2677","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5055097686","display_name":"Erke Ar\u0131ba\u015f","orcid":"https://orcid.org/0000-0003-2780-6621","observed_orcids":["https://orcid.org/0000-0003-2780-6621"]},"institutions":[{"id":"https://openalex.org/I48912391","display_name":"Istanbul Technical University","ror":"https://ror.org/059636586","country_code":"TR","type":"education","lineage":["https://openalex.org/I48912391"]}],"countries":["TR"],"is_corresponding":false,"raw_author_name":"Ar\u0131ba\u015f, Erke","raw_affiliation_strings":["Istanbul Technical University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Istanbul Technical University","institution_ids":["https://openalex.org/I48912391"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001750705","display_name":"Melisa G\u00fcler","orcid":"https://orcid.org/0009-0004-4233-0603","observed_orcids":["https://orcid.org/0009-0004-4233-0603"]},"institutions":[{"id":"https://openalex.org/I48912391","display_name":"Istanbul Technical University","ror":"https://ror.org/059636586","country_code":"TR","type":"education","lineage":["https://openalex.org/I48912391"]}],"countries":["TR"],"is_corresponding":false,"raw_author_name":"G\u00fcler, Melisa","raw_affiliation_strings":["Istanbul Technical University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Istanbul Technical University","institution_ids":["https://openalex.org/I48912391"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I48912391"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9570000171661377,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9570000171661377,"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"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.029100000858306885,"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":0.002300000051036477,"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/llama-2","display_name":"Llama 2","score":0.9399999976158142},{"id":"https://openalex.org/keywords/llama-3","display_name":"Llama 3","score":1.0},{"id":"https://openalex.org/keywords/model-compression","display_name":"model compression","score":0.6330000162124634},{"id":"https://openalex.org/keywords/weight-pruning","display_name":"weight pruning","score":0.7710000276565552},{"id":"https://openalex.org/keywords/lora","display_name":"LoRa","score":0.9539999961853027},{"id":"https://openalex.org/keywords/qlora","display_name":"QLoRA","score":0.9869999885559082},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.7919999957084656},{"id":"https://openalex.org/keywords/resource-constrained-devices","display_name":"resource-constrained devices","score":0.847000002861023},{"id":"https://openalex.org/keywords/memory-efficiency","display_name":"memory efficiency","score":0.6320000290870667},{"id":"https://openalex.org/keywords/inference-latency","display_name":"inference latency","score":0.7799999713897705},{"id":"https://openalex.org/keywords/parameter-efficient-fine-tuning","display_name":"parameter-efficient fine-tuning","score":0.6000000238418579}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7735000252723694},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.6007999777793884},{"id":"https://openalex.org/C20556612","wikidata":"https://www.wikidata.org/wiki/Q4469374","display_name":"Volume (thermodynamics)","level":2,"score":0.5511999726295471},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.519599974155426},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.5123000144958496},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.4180000126361847},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3912000060081482},{"id":"https://openalex.org/C94835093","wikidata":"https://www.wikidata.org/wiki/Q3113333","display_name":"Data compression ratio","level":5,"score":0.3880000114440918},{"id":"https://openalex.org/C25797200","wikidata":"https://www.wikidata.org/wiki/Q828137","display_name":"Compression ratio","level":3,"score":0.3725000023841858},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.37119999527931213},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.37119999527931213},{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.35120001435279846},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.33550000190734863},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.32199999690055847},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.31439998745918274},{"id":"https://openalex.org/C194739806","wikidata":"https://www.wikidata.org/wiki/Q66221","display_name":"Computer data storage","level":2,"score":0.31040000915527344},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.28360000252723694},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.28200000524520874},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.2806999981403351},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.2768999934196472},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2703999876976013},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2678000032901764},{"id":"https://openalex.org/C121838778","wikidata":"https://www.wikidata.org/wiki/Q7619924","display_name":"Storage efficiency","level":2,"score":0.26510000228881836},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.26249998807907104},{"id":"https://openalex.org/C2987595161","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Optimization algorithm","level":2,"score":0.25600001215934753}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:oai.as-proceeding.com:article/2810","doi":null,"is_oa":true,"landing_page_url":"https://as-proceeding.com/index.php/ijanser/article/view/2810","pdf_url":"https://as-proceeding.com/index.php/ijanser/article/download/2810/2677","source":{"id":"https://openalex.org/S4387288395","display_name":"International Journal of Advanced Natural Sciences and Engineering Researches","issn_l":"2980-0811","issn":["2980-0811"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2980-0811","raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"pmh:oai:oai.as-proceeding.com:article/2810","doi":null,"is_oa":true,"landing_page_url":"https://as-proceeding.com/index.php/ijanser/article/view/2810","pdf_url":"https://as-proceeding.com/index.php/ijanser/article/download/2810/2677","source":{"id":"https://openalex.org/S4387288395","display_name":"International Journal of Advanced Natural Sciences and Engineering Researches","issn_l":"2980-0811","issn":["2980-0811"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2980-0811","raw_type":"info:eu-repo/semantics/article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7217744112.pdf","grobid_xml":"https://content.openalex.org/works/W7217744112.grobid-xml"},"referenced_works_count":5,"referenced_works":["https://openalex.org/W3030163527","https://openalex.org/W4313484599","https://openalex.org/W4322718191","https://openalex.org/W4384918448","https://openalex.org/W7133233472"],"related_works":["https://openalex.org/W7168082978","https://openalex.org/W4391375196","https://openalex.org/W4407124521","https://openalex.org/W4414019041","https://openalex.org/W7143343316","https://openalex.org/W4396786590","https://openalex.org/W4404782212","https://openalex.org/W4412676021","https://openalex.org/W7103754027","https://openalex.org/W4391272673"],"abstract_inverted_index":{"Data":[0],"exists":[1],"in":[2,100,172,205,254],"different":[3],"sizes":[4,80],"according":[5],"to":[6,56,227,261],"the":[7,24,61,86,107,116,123,138,144,153,159,190,195,202,217,222,233,245,251,255,259,266,271,278],"needs":[8],"and":[9,36,81,104,109,122,148,177,199,216,235,238],"usage":[10],"area.":[11],"The":[12,91,179],"need":[13,63,246],"for":[14,66,72,137,247],"these":[15,28],"storage":[16,32,68,176],"areas":[17],"is":[18,54,113,134,219,280],"increasing":[19,25],"day":[20],"by":[21,186,249],"day.":[22],"However,":[23],"size":[26,268],"of":[27,88,93,165,258,274,285],"models":[29],"significantly":[30],"increases":[31],"requirements,":[33],"memory":[34,240],"usage,":[35],"inference":[37],"latency,":[38],"limiting":[39],"applicability,":[40],"especially":[41,171],"on":[42,232],"resourceconstrained":[43],"devices":[44,173],"(e.g.,":[45],"mobile,":[46],"embedded":[47],"systems).":[48],"By":[49],"2025,":[50],"global":[51],"data":[52],"volume":[53],"expected":[55],"reach":[57],"175":[58],"zettabytes,":[59],"underscoring":[60],"growing":[62],"not":[64,135],"only":[65],"efficient":[67,110],"solutions":[69],"but":[70],"also":[71],"effective":[73],"model":[74,79,89,218,260,267,279],"optimization":[75],"techniques":[76],"[1].":[77],"Increasing":[78],"evolving":[82],"computational":[83],"requirements":[84],"reveal":[85],"importance":[87],"compression.":[90],"choice":[92],"compression":[94,111,117],"algorithm":[95,132],"plays":[96],"a":[97],"decisive":[98],"role":[99],"balancing":[101],"efficiency,":[102],"speed,":[103],"accuracy.":[105],"If":[106],"correct":[108],"technique":[112,182],"chosen,":[114],"both":[115],"ratio":[118],"can":[119,126,140,167,183,210],"be":[120,127,184,211],"increased":[121],"processing":[124],"time":[125],"reduced,":[128],"while":[129,269],"choosing":[130],"an":[131],"that":[133,161,209],"suitable":[136],"purpose":[139],"distract":[141],"us":[142],"from":[143],"target.":[145],"When":[146],"llama-2":[147],"llama-3,":[149],"which":[150],"are":[151,157,214,292],"among":[152],"large":[154,288],"language":[155],"models,":[156,237,289],"examined,":[158],"fact":[160],"they":[162],"have":[163],"millions":[164],"parameters":[166],"cause":[168],"serious":[169],"problems,":[170],"with":[174,282],"limited":[175],"performance.":[178],"appropriate":[180],"compaction":[181],"selected":[185],"taking":[187],"into":[188],"account":[189],"characteristics":[191],"as":[192,194],"well":[193],"overall":[196],"performance":[197],"metrics":[198],"size.":[200],"With":[201],"SparseGPT":[203],"used":[204],"this":[206],"study,":[207],"weights":[208,253],"considered":[212],"unnecessary":[213],"pruned":[215],"reduced.":[220],"Using":[221],"SparseGPT-based":[223],"pruning":[224],"method,":[225],"up":[226],"5060%":[228],"sparsity":[229],"was":[230],"applied":[231],"LLaMA-2":[234],"LLaMA-3":[236],"significant":[239],"savings":[241],"were":[242],"achieved":[243],"without":[244],"retraining":[248,287],"setting":[250],"insignificant":[252],"parameter":[256],"space":[257],"zero.":[262],"This":[263],"approach":[264],"reduces":[265],"eliminating":[270],"costly":[272],"requirement":[273],"full":[275],"retraining.":[276],"Then":[277],"customized":[281],"lora/qlora.":[283],"Instead":[284],"completely":[286],"additional":[290],"adjustments":[291],"made":[293],"through":[294],"adapters.":[295]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-03T00:00:00"}