{"id":"https://openalex.org/W4412673541","doi":"https://doi.org/10.1145/3731120.3744595","title":"Comparing Retrieval-Augmentation and Parameter-Efficient Fine-Tuning for Privacy-Preserving Personalization of Large Language Models","display_name":"Comparing Retrieval-Augmentation and Parameter-Efficient Fine-Tuning for Privacy-Preserving Personalization of Large Language Models","publication_year":2025,"publication_date":"2025-07-18","ids":{"openalex":"https://openalex.org/W4412673541","doi":"https://doi.org/10.1145/3731120.3744595"},"language":"en","primary_location":{"id":"doi:10.1145/3731120.3744595","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3731120.3744595","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2025 International ACM SIGIR Conference on Innovative Concepts and Theories in Information Retrieval (ICTIR)","raw_type":"proceedings-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/A5082594835","display_name":"Alireza Salemi","orcid":"https://orcid.org/0009-0006-1937-2615"},"institutions":[{"id":"https://openalex.org/I24603500","display_name":"University of Massachusetts Amherst","ror":"https://ror.org/0072zz521","country_code":"US","type":"education","lineage":["https://openalex.org/I24603500"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Alireza Salemi","raw_affiliation_strings":["University of Massachusetts Amherst, Amherst, MA, USA"],"affiliations":[{"raw_affiliation_string":"University of Massachusetts Amherst, Amherst, MA, USA","institution_ids":["https://openalex.org/I24603500"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101457713","display_name":"Hamed Zamani","orcid":"https://orcid.org/0000-0002-0800-3340"},"institutions":[{"id":"https://openalex.org/I24603500","display_name":"University of Massachusetts Amherst","ror":"https://ror.org/0072zz521","country_code":"US","type":"education","lineage":["https://openalex.org/I24603500"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Hamed Zamani","raw_affiliation_strings":["University of Massachusetts Amherst, Amherst, MA, USA"],"affiliations":[{"raw_affiliation_string":"University of Massachusetts Amherst, Amherst, MA, USA","institution_ids":["https://openalex.org/I24603500"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5082594835"],"corresponding_institution_ids":["https://openalex.org/I24603500"],"apc_list":null,"apc_paid":null,"fwci":9.7011,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.97713833,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"286","last_page":"296"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9973000288009644,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.9973000288009644,"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/T10028","display_name":"Topic Modeling","score":0.9972000122070312,"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/T11719","display_name":"Data Quality and Management","score":0.9940999746322632,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8333138823509216},{"id":"https://openalex.org/keywords/personalization","display_name":"Personalization","score":0.8315317630767822},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4320879578590393},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.35050085186958313},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2413083016872406},{"id":"https://openalex.org/keywords/world-wide-web","display_name":"World Wide Web","score":0.23468264937400818}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8333138823509216},{"id":"https://openalex.org/C183003079","wikidata":"https://www.wikidata.org/wiki/Q1000371","display_name":"Personalization","level":2,"score":0.8315317630767822},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4320879578590393},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.35050085186958313},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2413083016872406},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"score":0.23468264937400818}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3731120.3744595","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3731120.3744595","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2025 International ACM SIGIR Conference on Innovative Concepts and Theories in Information Retrieval (ICTIR)","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":32,"referenced_works":["https://openalex.org/W2029604041","https://openalex.org/W2051780479","https://openalex.org/W2064173066","https://openalex.org/W2076006872","https://openalex.org/W2559655401","https://openalex.org/W2587741066","https://openalex.org/W2972203331","https://openalex.org/W2979826702","https://openalex.org/W3102516861","https://openalex.org/W3155807546","https://openalex.org/W3156789018","https://openalex.org/W3169283738","https://openalex.org/W3190590916","https://openalex.org/W3212555993","https://openalex.org/W4229065825","https://openalex.org/W4235853093","https://openalex.org/W4283157527","https://openalex.org/W4307079201","https://openalex.org/W4317898419","https://openalex.org/W4384774018","https://openalex.org/W4385570777","https://openalex.org/W4385570815","https://openalex.org/W4385573236","https://openalex.org/W4385688545","https://openalex.org/W4386728881","https://openalex.org/W4391095130","https://openalex.org/W4396723585","https://openalex.org/W4396844090","https://openalex.org/W4400525546","https://openalex.org/W4401042391","https://openalex.org/W4402671806","https://openalex.org/W6927703899"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2109940557","https://openalex.org/W2466832359","https://openalex.org/W4391210591","https://openalex.org/W1582019636","https://openalex.org/W2390279801","https://openalex.org/W1499005795","https://openalex.org/W4366999490"],"abstract_inverted_index":{"Despite":[0],"its":[1],"substantial":[2],"impact":[3],"on":[4,112],"various":[5],"search,":[6],"recommendation,":[7],"and":[8,86,93,116,122,165],"question":[9],"answering":[10],"tasks,":[11],"privacy-preserving":[12],"methods":[13,119],"for":[14,79,83,178],"personalizing":[15],"large":[16],"language":[17],"models":[18],"(LLMs)":[19],"have":[20],"received":[21],"relatively":[22],"limited":[23,182],"exploration.":[24],"There":[25],"is":[26,175,192],"one":[27],"primary":[28],"approach":[29,59],"in":[30,147],"this":[31],"area":[32],"through":[33,68],"retrieval-augmented":[34],"generation":[35],"(RAG),":[36],"which":[37],"generates":[38],"personalized":[39,149],"outputs":[40],"by":[41],"enriching":[42],"the":[43,50,75,105,142,159,166],"input":[44],"prompt":[45],"with":[46,132,181],"information":[47],"retrieved":[48],"from":[49,104],"user's":[51],"personal":[52,183],"data.":[53],"This":[54,72,170],"paper":[55,73],"studies":[56],"an":[57,88],"orthogonal":[58],"to":[60],"RAG":[61,131,174],"that":[62,173],"involves":[63],"learning":[64],"user-dependent":[65],"LLM":[66,84],"parameters":[67],"parameter-efficient":[69],"fine-tuning":[70],"(PEFT).":[71],"presents":[74],"first":[76],"systematic":[77],"study":[78],"exploration":[80],"of":[81,100,139,144,161,168],"PEFT":[82,185],"personalization":[85,118],"provides":[87],"extensive":[89],"comparisons":[90],"between":[91,158],"RAG-":[92,115],"PEFT-based":[94,117],"solutions,":[95],"across":[96],"a":[97,136,155],"broad":[98],"set":[99],"seven":[101],"diverse":[102],"datasets":[103],"LaMP":[106],"benchmark.":[107],"Our":[108],"results":[109],"demonstrate":[110],"that,":[111],"average,":[113],"both":[114],"yield":[120],"14.92%":[121],"1.07%":[123],"improvements":[124],"over":[125],"non-personalized":[126],"LLMs,":[127],"respectively.":[128],"When":[129],"combining":[130],"PEFT,":[133],"we":[134,153],"observe":[135],"further":[137],"improvement":[138],"15.98%,":[140],"highlighting":[141],"effectiveness":[143,167],"their":[145],"integration":[146],"enhancing":[148],"text":[150],"generation.":[151],"Additionally,":[152],"identify":[154],"positive":[156],"correlation":[157],"amount":[160],"user":[162],"data":[163,191],"available":[164],"PEFT.":[169],"finding":[171],"suggests":[172],"particularly":[176],"beneficial":[177],"cold-start":[179],"users--users":[180],"data--while":[184],"performs":[186],"better":[187],"when":[188],"more":[189],"user-specific":[190],"available.":[193]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2}],"updated_date":"2026-04-02T15:55:50.835912","created_date":"2025-10-10T00:00:00"}