{"id":"https://openalex.org/W4415546061","doi":"https://doi.org/10.1145/3746262.3761973","title":"SLMQuant: Benchmarking Small Language Model Quantization for Practical Deployment","display_name":"SLMQuant: Benchmarking Small Language Model Quantization for Practical Deployment","publication_year":2025,"publication_date":"2025-10-25","ids":{"openalex":"https://openalex.org/W4415546061","doi":"https://doi.org/10.1145/3746262.3761973"},"language":null,"primary_location":{"id":"doi:10.1145/3746262.3761973","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746262.3761973","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 3rd International Workshop on Rich Media With Generative AI","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3746262.3761973","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5070352862","display_name":"J.J. Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210100255","display_name":"Beijing Academy of Artificial Intelligence","ror":"https://ror.org/016a74861","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210100255"]},{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiacheng Wang","raw_affiliation_strings":["SKLCCSE School of Artificial Intelligence, Beihang University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0006-5393-0000","affiliations":[{"raw_affiliation_string":"SKLCCSE School of Artificial Intelligence, Beihang University, Beijing, China","institution_ids":["https://openalex.org/I4210100255","https://openalex.org/I82880672"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101221475","display_name":"Yejun Zeng","orcid":"https://orcid.org/0009-0006-6445-0571"},"institutions":[{"id":"https://openalex.org/I4210100255","display_name":"Beijing Academy of Artificial Intelligence","ror":"https://ror.org/016a74861","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210100255"]},{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yejun Zeng","raw_affiliation_strings":["SKLCCSE School of Artificial Intelligence, Beihang University, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0006-6445-0571","affiliations":[{"raw_affiliation_string":"SKLCCSE School of Artificial Intelligence, Beihang University, Beijing, China","institution_ids":["https://openalex.org/I4210100255","https://openalex.org/I82880672"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039703932","display_name":"Jinyang Guo","orcid":"https://orcid.org/0000-0003-1956-3367"},"institutions":[{"id":"https://openalex.org/I4210100255","display_name":"Beijing Academy of Artificial Intelligence","ror":"https://ror.org/016a74861","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210100255"]},{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jinyang Guo","raw_affiliation_strings":["SKLCCSE School of Artificial Intelligence, Beihang University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-1956-3367","affiliations":[{"raw_affiliation_string":"SKLCCSE School of Artificial Intelligence, Beihang University, Beijing, China","institution_ids":["https://openalex.org/I4210100255","https://openalex.org/I82880672"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030290826","display_name":"Yuqing Ma","orcid":"https://orcid.org/0000-0003-1936-9396"},"institutions":[{"id":"https://openalex.org/I4210100255","display_name":"Beijing Academy of Artificial Intelligence","ror":"https://ror.org/016a74861","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210100255"]},{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuqing Ma","raw_affiliation_strings":["SKLCCSE School of Artificial Intelligence, Beihang University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-1936-9396","affiliations":[{"raw_affiliation_string":"SKLCCSE School of Artificial Intelligence, Beihang University, Beijing, China","institution_ids":["https://openalex.org/I4210100255","https://openalex.org/I82880672"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014870180","display_name":"Aishan Liu","orcid":"https://orcid.org/0000-0002-4224-1318"},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Aishan Liu","raw_affiliation_strings":["SKLCCSE School of Computer Science and Engineering, Beihang University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-4224-1318","affiliations":[{"raw_affiliation_string":"SKLCCSE School of Computer Science and Engineering, Beihang University, Beijing, China","institution_ids":["https://openalex.org/I82880672"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024067284","display_name":"Xianglong Liu","orcid":"https://orcid.org/0000-0001-8425-4195"},"institutions":[{"id":"https://openalex.org/I82880672","display_name":"Beihang University","ror":"https://ror.org/00wk2mp56","country_code":"CN","type":"education","lineage":["https://openalex.org/I82880672"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xianglong Liu","raw_affiliation_strings":["SKLCCSE School of Computer Science and Engineering, Beihang University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-8425-4195","affiliations":[{"raw_affiliation_string":"SKLCCSE School of Computer Science and Engineering, Beihang University, Beijing, China","institution_ids":["https://openalex.org/I82880672"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.1569173,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2","last_page":"10"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9451000094413757,"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":0.9451000094413757,"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":0.9272000193595886,"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/quantization","display_name":"Quantization (signal processing)","score":0.7598000168800354},{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.7315000295639038},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.7132999897003174},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.47350001335144043},{"id":"https://openalex.org/keywords/edge-device","display_name":"Edge device","score":0.45170000195503235},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.4462999999523163}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.7598000168800354},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.7315000295639038},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.7132999897003174},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7010999917984009},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.47350001335144043},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.45170000195503235},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.4462999999523163},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.36910000443458557},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3522999882698059},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.29330000281333923},{"id":"https://openalex.org/C45493050","wikidata":"https://www.wikidata.org/wiki/Q7884934","display_name":"Unified Model","level":2,"score":0.2897999882698059},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.28929999470710754},{"id":"https://openalex.org/C199833920","wikidata":"https://www.wikidata.org/wiki/Q612536","display_name":"Vector quantization","level":2,"score":0.27300000190734863},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.2662999927997589},{"id":"https://openalex.org/C134537474","wikidata":"https://www.wikidata.org/wiki/Q17144832","display_name":"Naturalness","level":2,"score":0.2565000057220459}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3746262.3761973","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746262.3761973","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 3rd International Workshop on Rich Media With Generative AI","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2511.13023","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2511.13023","pdf_url":"https://arxiv.org/pdf/2511.13023","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"}],"best_oa_location":{"id":"doi:10.1145/3746262.3761973","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3746262.3761973","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 3rd International Workshop on Rich Media With Generative AI","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":4,"referenced_works":["https://openalex.org/W2998617917","https://openalex.org/W2998715470","https://openalex.org/W4290717053","https://openalex.org/W4402715871"],"related_works":[],"abstract_inverted_index":{"Despite":[0],"the":[1,60],"growing":[2],"interest":[3],"in":[4,29,100,152,164],"Small":[5],"Language":[6,14],"Models":[7,15],"(SLMs)":[8],"as":[9],"resource-efficient":[10],"alternatives":[11],"to":[12,25,41,71,111,115],"Large":[13],"(LLMs),":[16],"their":[17],"deployment":[18,148],"on":[19,89,149],"edge":[20,153],"devices":[21,151],"remains":[22],"challenging":[23],"due":[24,114],"unresolved":[26],"efficiency":[27,54],"gaps":[28],"model":[30],"compression.":[31,138],"While":[32],"quantization":[33,51,86,101,130],"has":[34],"proven":[35],"effective":[36,128],"for":[37,64,136,144,159],"LLMs,":[38],"its":[39],"applicability":[40],"SLMs":[42,97],"is":[43],"significantly":[44],"underexplored,":[45],"with":[46],"critical":[47,157],"questions":[48],"about":[49],"differing":[50],"bottlenecks":[52],"and":[53,80,98,120,131,155],"profiles.":[55],"This":[56],"paper":[57],"introduces":[58],"SLMQuant,":[59],"first":[61],"systematic":[62],"benchmark":[63],"evaluating":[65],"LLM":[66],"compression":[67],"techniques":[68,109],"when":[69],"applied":[70],"SLMs.":[72,90],"Through":[73],"comprehensive":[74],"multi-track":[75],"evaluations":[76],"across":[77],"diverse":[78],"architectures":[79],"tasks,":[81],"we":[82],"analyze":[83],"how":[84],"state-of-the-art":[85],"methods":[87],"perform":[88],"Our":[91],"findings":[92],"reveal":[93],"fundamental":[94],"disparities":[95],"between":[96],"LLMs":[99],"sensitivity,":[102],"demonstrating":[103],"that":[104],"direct":[105],"transfer":[106],"of":[107],"LLM-optimized":[108],"leads":[110],"suboptimal":[112],"results":[113],"SLMs'":[116],"unique":[117],"architectural":[118],"characteristics":[119],"training":[121],"dynamics.":[122],"We":[123],"identify":[124],"key":[125],"factors":[126],"governing":[127],"SLM":[129,147],"propose":[132],"actionable":[133],"design":[134],"principles":[135],"SLM-tailored":[137],"SLMQuant":[139],"establishes":[140],"a":[141],"foundational":[142],"framework":[143],"advancing":[145],"efficient":[146],"low-end":[150],"applications,":[154],"provides":[156],"insights":[158],"deploying":[160],"lightweight":[161],"language":[162],"models":[163],"resource-constrained":[165],"scenarios.":[166]},"counts_by_year":[],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2025-10-25T00:00:00"}