{"id":"https://openalex.org/W7133332521","doi":"https://doi.org/10.1109/tvlsi.2026.3666193","title":"DayPQ: Dynamic Layerwise Pruning and Quantization for LLM Inference Acceleration","display_name":"DayPQ: Dynamic Layerwise Pruning and Quantization for LLM Inference Acceleration","publication_year":2026,"publication_date":"2026-03-03","ids":{"openalex":"https://openalex.org/W7133332521","doi":"https://doi.org/10.1109/tvlsi.2026.3666193"},"language":null,"primary_location":{"id":"doi:10.1109/tvlsi.2026.3666193","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2026.3666193","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"},"type":"article","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/A5127883102","display_name":"Yongchao Xu","orcid":null},"institutions":[{"id":"https://openalex.org/I180726961","display_name":"Shenzhen University","ror":"https://ror.org/01vy4gh70","country_code":"CN","type":"education","lineage":["https://openalex.org/I180726961"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yongchao Xu","raw_affiliation_strings":["School of Integrated Circuits, Sun Yat-sen University, Shenzhen Campus, Shenzhen, China"],"raw_orcid":"https://orcid.org/0009-0000-4043-1655","affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Sun Yat-sen University, Shenzhen Campus, Shenzhen, China","institution_ids":["https://openalex.org/I180726961"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127975168","display_name":"Mingxuan Li","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mingxuan Li","raw_affiliation_strings":["School of Integrated Circuits, Peking University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128027426","display_name":"Ziying Zhuo","orcid":null},"institutions":[{"id":"https://openalex.org/I180726961","display_name":"Shenzhen University","ror":"https://ror.org/01vy4gh70","country_code":"CN","type":"education","lineage":["https://openalex.org/I180726961"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ziying Zhuo","raw_affiliation_strings":["School of Integrated Circuits, Sun Yat-sen University, Shenzhen Campus, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Sun Yat-sen University, Shenzhen Campus, Shenzhen, China","institution_ids":["https://openalex.org/I180726961"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114554075","display_name":"Qinzhe Zhi","orcid":"https://orcid.org/0009-0001-4866-7213"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qinzhe Zhi","raw_affiliation_strings":["School of Integrated Circuits, Peking University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Tianyu Jia","orcid":"https://orcid.org/0000-0002-4570-4613"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tianyu Jia","raw_affiliation_strings":["School of Integrated Circuits, Peking University, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-4570-4613","affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Peking University, Beijing, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5122916732","display_name":"Meiqi Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I180726961","display_name":"Shenzhen University","ror":"https://ror.org/01vy4gh70","country_code":"CN","type":"education","lineage":["https://openalex.org/I180726961"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Meiqi Wang","raw_affiliation_strings":["School of Integrated Circuits, Sun Yat-sen University, Shenzhen Campus, Shenzhen, China"],"raw_orcid":"https://orcid.org/0000-0001-9553-3640","affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Sun Yat-sen University, Shenzhen Campus, Shenzhen, China","institution_ids":["https://openalex.org/I180726961"]}]}],"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":0.20589611,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"34","issue":"6","first_page":"1712","last_page":"1722"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.1006999984383583,"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":0.1006999984383583,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.05739999935030937,"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":0.043800000101327896,"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/quantization","display_name":"Quantization (signal processing)","score":0.6861000061035156},{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.4821000099182129},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4650000035762787},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.41600000858306885},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.40639999508857727},{"id":"https://openalex.org/keywords/offset","display_name":"Offset (computer science)","score":0.39640000462532043},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.3752000033855438},{"id":"https://openalex.org/keywords/normalization","display_name":"Normalization (sociology)","score":0.37310001254081726},{"id":"https://openalex.org/keywords/logarithm","display_name":"Logarithm","score":0.35030001401901245}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.741100013256073},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.6861000061035156},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.536300003528595},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.4821000099182129},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4650000035762787},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.41600000858306885},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.40639999508857727},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.40610000491142273},{"id":"https://openalex.org/C175291020","wikidata":"https://www.wikidata.org/wiki/Q1156822","display_name":"Offset (computer science)","level":2,"score":0.39640000462532043},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.3752000033855438},{"id":"https://openalex.org/C136886441","wikidata":"https://www.wikidata.org/wiki/Q926129","display_name":"Normalization (sociology)","level":2,"score":0.37310001254081726},{"id":"https://openalex.org/C39927690","wikidata":"https://www.wikidata.org/wiki/Q11197","display_name":"Logarithm","level":2,"score":0.35030001401901245},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.32120001316070557},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.31790000200271606},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.3091000020503998},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.30329999327659607},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.30320000648498535},{"id":"https://openalex.org/C2221639","wikidata":"https://www.wikidata.org/wiki/Q2877","display_name":"Discrete cosine transform","level":3,"score":0.3019999861717224},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.2881999909877777},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.2833000123500824},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.27880001068115234},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.27399998903274536},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.2718999981880188},{"id":"https://openalex.org/C55439883","wikidata":"https://www.wikidata.org/wiki/Q360812","display_name":"Correctness","level":2,"score":0.2639999985694885},{"id":"https://openalex.org/C40760162","wikidata":"https://www.wikidata.org/wiki/Q10920295","display_name":"Voltage droop","level":4,"score":0.2606000006198883},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.25920000672340393},{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.2565999925136566},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.25459998846054077},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tvlsi.2026.3666193","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tvlsi.2026.3666193","pdf_url":null,"source":{"id":"https://openalex.org/S37538908","display_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","issn_l":"1063-8210","issn":["1063-8210","1557-9999"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Very Large Scale Integration (VLSI) Systems","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.5262514352798462,"display_name":"Affordable and clean energy"}],"awards":[{"id":"https://openalex.org/G1462451646","display_name":null,"funder_award_id":"62404256","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"deployment":[1],"of":[2,37,91,164,190],"the":[3,35,89,120,134,141,159,183],"prevailing":[4],"large":[5],"language":[6],"models":[7],"(LLMs)":[8],"on":[9],"resource-constrained":[10],"hardware":[11,74,135],"faces":[12],"critical":[13,53],"challenges":[14],"due":[15],"to":[16],"their":[17],"massive":[18],"computational":[19],"and":[20,24,55,76,84,115,130,145,174,211],"memory":[21],"requirements.":[22],"Pruning":[23],"quantization":[25,118],"(PQ),":[26],"as":[27],"two":[28],"predominant":[29],"techniques":[30],"for":[31,108],"LLM":[32,110],"acceleration,":[33],"face":[34],"challenge":[36],"fully":[38],"preserving":[39],"model":[40],"capabilities":[41],"while":[42],"achieving":[43,137],"significant":[44],"acceleration":[45,139],"in":[46],"practice.":[47],"Static":[48],"PQ":[49,68,87,152],"irreversibly":[50],"discards":[51],"potentially":[52],"parameters":[54],"degrades":[56],"precision":[57],"across":[58],"diverse":[59],"inputs,":[60],"which":[61],"must":[62],"be":[63],"compensated":[64],"through":[65],"fine-tuning.":[66],"Dynamic":[67],"preserves":[69],"accuracy":[70,143],"but":[71],"incurs":[72],"prohibitive":[73],"overhead":[75],"additional":[77,147],"latency":[78],"from":[79,158],"runtime":[80],"adaptations.":[81],"Achieving":[82],"low-latency":[83],"overhead-efficient":[85],"dynamic":[86,138,170],"requires":[88],"co-design":[90],"adaptive":[92,151],"compression":[93,179],"tightly":[94],"integrated":[95],"with":[96,124,140],"hardware.":[97],"In":[98],"this":[99],"work,":[100],"we":[101],"present":[102],"DayPQ,":[103],"an":[104,199],"algorithm-hardware":[105],"co-optimization":[106],"technique":[107],"efficient":[109],"inference:":[111],"self-adaptive":[112],"norm-based":[113],"pruning":[114],"bias-shifting":[116],"FP8":[117],"at":[119,133],"algorithm":[121],"level,":[122,136],"synergized":[123],"a":[125],"multiplication-free":[126],"shift\u2013add":[127],"PE":[128],"array":[129],"switchable":[131],"dataflow":[132],"negligible":[142],"loss":[144],"zero":[146],"latency.":[148],"DayPQ":[149,197],"enables":[150],"by":[153],"intrinsically":[154],"exploiting":[155],"normalization":[156],"operations":[157],"model\u2019s":[160],"inference":[161],"process":[162],"instead":[163],"extra":[165],"operations.":[166,192],"Simultaneously,":[167],"it":[168],"establishes":[169],"orthogonality":[171],"between":[172],"sparsity":[173],"quantization,":[175],"where":[176],"these":[177],"dual":[178],"mechanisms":[180],"synergistically":[181],"enhance":[182],"redundancy":[184],"reduction":[185],"beyond":[186],"mere":[187],"additive":[188],"effects":[189],"independent":[191],"Our":[193],"evaluation":[194],"shows":[195],"that":[196],"obtains":[198],"average":[200],"<inline-formula":[201,212],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[202,213],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">":[203,214],"<tex-math":[204,215],"notation=\"LaTeX\">$3.13\\times":[205],"$</tex-math>":[206,217],"</inline-formula>":[207,218],"higher":[208],"energy":[209],"efficiency":[210],"notation=\"LaTeX\">$2.03\\times":[216],"performance":[219],"boost":[220],"than":[221],"three":[222],"state-of-the-art":[223],"(SOTA)":[224],"accelerators.":[225]},"counts_by_year":[],"updated_date":"2026-06-03T06:16:58.514037","created_date":"2026-03-04T00:00:00"}