{"id":"https://openalex.org/W4412876835","doi":"https://doi.org/10.1145/3711896.3736919","title":"EARN: Efficient Inference Acceleration for LLM-based Generative Recommendation by Register Tokens","display_name":"EARN: Efficient Inference Acceleration for LLM-based Generative Recommendation by Register Tokens","publication_year":2025,"publication_date":"2025-08-03","ids":{"openalex":"https://openalex.org/W4412876835","doi":"https://doi.org/10.1145/3711896.3736919"},"language":"en","primary_location":{"id":"doi:10.1145/3711896.3736919","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3711896.3736919","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3711896.3736919","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 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3711896.3736919","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101250832","display_name":"Chaoqun Yang","orcid":null},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Chaoqun Yang","raw_affiliation_strings":["Tsinghua University, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083746705","display_name":"Xinyu Lin","orcid":"https://orcid.org/0000-0002-6931-3182"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Xinyu Lin","raw_affiliation_strings":["National University of Singapore, Singapore, Singapore"],"affiliations":[{"raw_affiliation_string":"National University of Singapore, Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100368524","display_name":"Wenjie Wang","orcid":"https://orcid.org/0000-0002-5199-1428"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenjie Wang","raw_affiliation_strings":["University of Science and Technology of China, Hefei, China"],"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China, Hefei, China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101599249","display_name":"Yongqi Li","orcid":"https://orcid.org/0000-0002-6932-4228"},"institutions":[{"id":"https://openalex.org/I14243506","display_name":"Hong Kong Polytechnic University","ror":"https://ror.org/0030zas98","country_code":"HK","type":"education","lineage":["https://openalex.org/I14243506"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yongqi Li","raw_affiliation_strings":["The Hong Kong Polytechnic University, Hong Kong, China"],"affiliations":[{"raw_affiliation_string":"The Hong Kong Polytechnic University, Hong Kong, China","institution_ids":["https://openalex.org/I14243506"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101791255","display_name":"Teng Sun","orcid":"https://orcid.org/0000-0003-0932-8910"},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Teng Sun","raw_affiliation_strings":["Shandong University, Qingdao, China"],"affiliations":[{"raw_affiliation_string":"Shandong University, Qingdao, China","institution_ids":["https://openalex.org/I154099455"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101500086","display_name":"Xianjing Han","orcid":"https://orcid.org/0000-0001-7867-3190"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Xianjing Han","raw_affiliation_strings":["National University of Singapore, Singapore, Singapore"],"affiliations":[{"raw_affiliation_string":"National University of Singapore, Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5089404640","display_name":"Tat\u2010Seng Chua","orcid":"https://orcid.org/0000-0001-6097-7807"},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Tat-Seng Chua","raw_affiliation_strings":["National University of Singapore, Singapore, Singapore"],"affiliations":[{"raw_affiliation_string":"National University of Singapore, Singapore, Singapore","institution_ids":["https://openalex.org/I165932596"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5101250832"],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":3.2508,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.92816152,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"3483","last_page":"3494"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10203","display_name":"Recommender Systems and Techniques","score":0.9846000075340271,"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"}},"topics":[{"id":"https://openalex.org/T10203","display_name":"Recommender Systems and Techniques","score":0.9846000075340271,"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"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9832000136375427,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.9830999970436096,"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/register","display_name":"Register (sociolinguistics)","score":0.7747560143470764},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.772915244102478},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7673947811126709},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.7635475993156433},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.6416870355606079},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4964907765388489},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.47693365812301636},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3427696228027344},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.33614566922187805}],"concepts":[{"id":"https://openalex.org/C2779235478","wikidata":"https://www.wikidata.org/wiki/Q286576","display_name":"Register (sociolinguistics)","level":2,"score":0.7747560143470764},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.772915244102478},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7673947811126709},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.7635475993156433},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.6416870355606079},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4964907765388489},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.47693365812301636},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3427696228027344},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.33614566922187805},{"id":"https://openalex.org/C74650414","wikidata":"https://www.wikidata.org/wiki/Q11397","display_name":"Classical mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3711896.3736919","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3711896.3736919","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3711896.3736919","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 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3711896.3736919","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3711896.3736919","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3711896.3736919","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 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320320671","display_name":"National Research Foundation","ror":"https://ror.org/05s0g1g46"},{"id":"https://openalex.org/F4320320709","display_name":"National Research Foundation Singapore","ror":"https://ror.org/03cpyc314"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412876835.pdf","grobid_xml":"https://content.openalex.org/works/W4412876835.grobid-xml"},"referenced_works_count":27,"referenced_works":["https://openalex.org/W3156844209","https://openalex.org/W3214693004","https://openalex.org/W4285247752","https://openalex.org/W4366330736","https://openalex.org/W4376122599","https://openalex.org/W4384655811","https://openalex.org/W4386187806","https://openalex.org/W4387848745","https://openalex.org/W4389519226","https://openalex.org/W4389523718","https://openalex.org/W4389524473","https://openalex.org/W4395444087","https://openalex.org/W4396606321","https://openalex.org/W4400531852","https://openalex.org/W4400606525","https://openalex.org/W4400909953","https://openalex.org/W4401834466","https://openalex.org/W4401863553","https://openalex.org/W4401863964","https://openalex.org/W4402683955","https://openalex.org/W4403221853","https://openalex.org/W4404639422","https://openalex.org/W4411119659","https://openalex.org/W6612746128","https://openalex.org/W6851812595","https://openalex.org/W6852797925","https://openalex.org/W6862154733"],"related_works":["https://openalex.org/W4365211920","https://openalex.org/W3014948380","https://openalex.org/W4391584540","https://openalex.org/W4380551139","https://openalex.org/W4317695495","https://openalex.org/W4395044357","https://openalex.org/W4287117424","https://openalex.org/W4387506531","https://openalex.org/W2087346071","https://openalex.org/W2967848559"],"abstract_inverted_index":{"Large":[0],"Language":[1],"Model-based":[2],"generative":[3],"recommendation":[4,42],"(LLMRec)":[5],"has":[6],"achieved":[7],"notable":[8],"success,":[9],"but":[10],"it":[11],"suffers":[12],"from":[13],"high":[14,84],"inference":[15,114],"latency":[16],"due":[17],"to":[18,121,160],"massive":[19],"computational":[20],"overhead":[21],"and":[22,86,99,151,163],"memory":[23],"pressure":[24],"of":[25,58,102],"KV":[26,29,165],"Cache.":[27],"Existing":[28],"Cache":[30,166],"reduction":[31,167],"methods":[32,150],"face":[33],"critical":[34],"limitations:":[35],"cache":[36],"compression":[37,49],"offers":[38],"marginal":[39],"acceleration":[40],"given":[41],"tasks'":[43],"short":[44],"decoding":[45],"steps,":[46],"while":[47,80],"prompt":[48],"risks":[50],"discarding":[51],"vital":[52],"interaction":[53],"history.":[54],"Through":[55],"systematic":[56],"analysis":[57],"attention":[59,70,89,93],"patterns":[60,79],"in":[61,139,182],"LLMRec,":[62,183],"we":[63,109],"uncover":[64],"two":[65,148,152],"pivotal":[66],"insights:":[67],"1)":[68],"layer-wise":[69],"sparsity":[71],"inversion":[72],"where":[73,92],"early":[74,119],"layers":[75,82,120],"retain":[76],"dense":[77],"informative":[78],"later":[81],"exhibit":[83],"redundancy,":[85],"2)":[87],"dual":[88],"sinks":[90],"phenomenon":[91],"scores":[94],"concentrate":[95],"on":[96,136,145],"both":[97],"head":[98],"tail":[100],"tokens":[101,126,138],"input":[103,130],"sequences.":[104],"Motivated":[105],"by":[106],"these":[107,137],"insights,":[108],"propose":[110],"EARN,":[111],"an":[112],"efficient":[113],"framework":[115],"that":[116],"leverages":[117],"the":[118,129,140,172,179],"compress":[122],"information":[123],"into":[124],"register":[125],"placed":[127],"at":[128],"sequence":[131],"boundaries,":[132],"then":[133],"focuses":[134],"solely":[135],"subsequent":[141],"layers.":[142],"Extensive":[143],"experiments":[144],"three":[146],"datasets,":[147],"LLMRec":[149],"LLM":[153],"architectures":[154],"demonstrate":[155],"EARN's":[156],"superiority,":[157],"achieving":[158],"up":[159],"3.79x":[161],"speedup":[162],"80.8%":[164],"with":[168],"better":[169],"accuracy":[170],"than":[171],"general":[173],"finetuning":[174],"approach.":[175],"Our":[176],"work":[177],"bridges":[178],"efficiency-effectiveness":[180],"gap":[181],"offering":[184],"practical":[185],"deployment":[186],"advantages":[187],"for":[188],"industrial":[189],"scenarios.":[190]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}