{"id":"https://openalex.org/W7124168798","doi":"https://doi.org/10.1109/icpads67057.2025.11322900","title":"PQC-LLM: Post-Quantization Delta Compression for LLMs","display_name":"PQC-LLM: Post-Quantization Delta Compression for LLMs","publication_year":2025,"publication_date":"2025-12-14","ids":{"openalex":"https://openalex.org/W7124168798","doi":"https://doi.org/10.1109/icpads67057.2025.11322900"},"language":null,"primary_location":{"id":"doi:10.1109/icpads67057.2025.11322900","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icpads67057.2025.11322900","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 31th International Conference on Parallel and Distributed Systems (ICPADS)","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/A5122993937","display_name":"Yujin Zhong","orcid":null},"institutions":[{"id":"https://openalex.org/I36399199","display_name":"Nanjing University of Science and Technology","ror":"https://ror.org/00xp9wg62","country_code":"CN","type":"education","lineage":["https://openalex.org/I36399199"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yujin Zhong","raw_affiliation_strings":["School of Computer Science and Technology, Nanjing University of Science and Technology,Nanjing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Technology, Nanjing University of Science and Technology,Nanjing,China","institution_ids":["https://openalex.org/I36399199"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090690232","display_name":"Fei Wu","orcid":"https://orcid.org/0000-0002-9366-2374"},"institutions":[{"id":"https://openalex.org/I36399199","display_name":"Nanjing University of Science and Technology","ror":"https://ror.org/00xp9wg62","country_code":"CN","type":"education","lineage":["https://openalex.org/I36399199"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chao Wu","raw_affiliation_strings":["School of Computer Science and Technology, Nanjing University of Science and Technology,Nanjing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Technology, Nanjing University of Science and Technology,Nanjing,China","institution_ids":["https://openalex.org/I36399199"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5116859576","display_name":"Cheng Ji","orcid":null},"institutions":[{"id":"https://openalex.org/I36399199","display_name":"Nanjing University of Science and Technology","ror":"https://ror.org/00xp9wg62","country_code":"CN","type":"education","lineage":["https://openalex.org/I36399199"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cheng Ji","raw_affiliation_strings":["School of Computer Science and Technology, Nanjing University of Science and Technology,Nanjing,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Technology, Nanjing University of Science and Technology,Nanjing,China","institution_ids":["https://openalex.org/I36399199"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I36399199"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.63030767,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.26739999651908875,"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":0.26739999651908875,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.18070000410079956,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T14347","display_name":"Big Data and Digital Economy","score":0.13179999589920044,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.6865000128746033},{"id":"https://openalex.org/keywords/lossy-compression","display_name":"Lossy compression","score":0.6431999802589417},{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.6151000261306763},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.5491999983787537},{"id":"https://openalex.org/keywords/tile","display_name":"Tile","score":0.5343000292778015},{"id":"https://openalex.org/keywords/compression","display_name":"Compression (physics)","score":0.4462999999523163},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.4431999921798706},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.38670000433921814}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6865000128746033},{"id":"https://openalex.org/C165021410","wikidata":"https://www.wikidata.org/wiki/Q55564","display_name":"Lossy compression","level":2,"score":0.6431999802589417},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.6151000261306763},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.5491999983787537},{"id":"https://openalex.org/C2780728851","wikidata":"https://www.wikidata.org/wiki/Q468402","display_name":"Tile","level":2,"score":0.5343000292778015},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.48820000886917114},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.4462999999523163},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.4431999921798706},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.38670000433921814},{"id":"https://openalex.org/C199833920","wikidata":"https://www.wikidata.org/wiki/Q612536","display_name":"Vector quantization","level":2,"score":0.358599990606308},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3400000035762787},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.32510000467300415},{"id":"https://openalex.org/C2777027219","wikidata":"https://www.wikidata.org/wiki/Q1284190","display_name":"Constant (computer programming)","level":2,"score":0.3246000111103058},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.30640000104904175},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3061000108718872},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.2985000014305115},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.2976999878883362},{"id":"https://openalex.org/C13481523","wikidata":"https://www.wikidata.org/wiki/Q412438","display_name":"Image compression","level":4,"score":0.28859999775886536},{"id":"https://openalex.org/C2776291640","wikidata":"https://www.wikidata.org/wiki/Q2912517","display_name":"Value (mathematics)","level":2,"score":0.2680000066757202},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.2635999917984009},{"id":"https://openalex.org/C5072461","wikidata":"https://www.wikidata.org/wiki/Q49506","display_name":"Delta","level":2,"score":0.26350000500679016},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.2603999972343445},{"id":"https://openalex.org/C94835093","wikidata":"https://www.wikidata.org/wiki/Q3113333","display_name":"Data compression ratio","level":5,"score":0.26019999384880066},{"id":"https://openalex.org/C134765980","wikidata":"https://www.wikidata.org/wiki/Q879126","display_name":"Bitwise operation","level":2,"score":0.2590000033378601},{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.2538999915122986}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icpads67057.2025.11322900","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icpads67057.2025.11322900","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 31th International Conference on Parallel and Distributed Systems (ICPADS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W2052440657","https://openalex.org/W2162752393","https://openalex.org/W2303422468","https://openalex.org/W2486202470","https://openalex.org/W2765756720","https://openalex.org/W2982317364","https://openalex.org/W3045703362","https://openalex.org/W3048079516","https://openalex.org/W4226087293","https://openalex.org/W4289441319","https://openalex.org/W4316661233","https://openalex.org/W4386290345","https://openalex.org/W4405907026"],"related_works":[],"abstract_inverted_index":{"The":[0],"massive":[1],"parameter":[2],"scale":[3],"and":[4,19,114],"substantial":[5],"storage":[6,144],"requirements":[7,145],"of":[8,88,116,146],"large":[9],"language":[10],"models":[11],"(LLMs)":[12],"hinder":[13],"their":[14],"deployment":[15],"on":[16,53,125],"edge":[17],"devices":[18],"in":[20],"real-world":[21],"applications,":[22],"even":[23],"after":[24],"quantization.":[25],"To":[26,71],"address":[27],"this":[28],"challenge,":[29],"we":[30,121],"propose":[31],"PQC-LLM,":[32],"a":[33,68,77,102],"Post-Quantization":[34],"delta":[35,123],"Compression":[36],"scheme":[37],"for":[38],"LLMs.":[39],"PQC-LLM":[40,75,140],"is":[41],"motivated":[42],"by":[43,82],"two":[44],"key":[45],"observations:":[46],"(1)":[47],"quantized":[48,147],"LLMs":[49,148],"still":[50],"exhibit":[51],"redundancy":[52],"the":[54,73,85,92,96,110,117,129,133,143],"least":[55,86],"significant":[56],"bits":[57],"(LSBs);":[58],"(2)":[59],"State-of-the-art":[60],"(SOTA)":[61],"quantization":[62],"techniques":[63],"could":[64],"only":[65],"converge":[66],"to":[67,108],"local":[69],"optimum.":[70],"exploit":[72],"potential,":[74],"performs":[76],"tile-wise":[78],"deterministic":[79],"bit":[80,98,112],"replacement":[81],"uniformly":[83],"substituting":[84],"significant$N$bits":[87],"all":[89],"weights":[90],"within":[91],"same":[93],"tile":[94,127],"with":[95,149],"constant":[97],"value.":[99,135],"We":[100],"employ":[101],"network":[103],"architecture":[104],"search":[105],"(NAS)":[106],"mechanism":[107],"determine":[109],"optimal":[111],"length":[113],"value":[115],"uniform":[118],"replacement.":[119],"Subsequently,":[120],"perform":[122],"compression":[124],"each":[126],"using":[128],"first":[130],"weight":[131],"as":[132],"base":[134],"Extensive":[136],"experiments":[137],"show":[138],"that,":[139],"effectively":[141],"reduces":[142],"negligible":[150],"lossy":[151],"or":[152],"improved":[153],"task":[154],"accuracy.":[155]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-01-15T00:00:00"}