{"id":"https://openalex.org/W4409671299","doi":"https://doi.org/10.1145/3696410.3714817","title":"Automatic Instruction Data Selection for Large Language Models via Uncertainty-Aware Influence Maximization","display_name":"Automatic Instruction Data Selection for Large Language Models via Uncertainty-Aware Influence Maximization","publication_year":2025,"publication_date":"2025-04-22","ids":{"openalex":"https://openalex.org/W4409671299","doi":"https://doi.org/10.1145/3696410.3714817"},"language":"en","primary_location":{"id":"doi:10.1145/3696410.3714817","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3696410.3714817","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3696410.3714817","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Web Conference 2025","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/10.1145/3696410.3714817","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5027644323","display_name":"Jindong Han","orcid":"https://orcid.org/0000-0002-1542-6149"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":true,"raw_author_name":"Jindong Han","raw_affiliation_strings":["The Hong Kong University of Science and Technology, Hong Kong, China"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology, Hong Kong, China","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100458897","display_name":"Hao Liu","orcid":"https://orcid.org/0000-0003-4271-1567"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao Liu","raw_affiliation_strings":["The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China and The Hong Kong University of Science and Technology, Hong Kong, China"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology (Guangzhou), Guangzhou, China and The Hong Kong University of Science and Technology, Hong Kong, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114948429","display_name":"Jun Fang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jun Fang","raw_affiliation_strings":["Didichuxing Co. Ltd, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Didichuxing Co. Ltd, Beijing, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020921636","display_name":"Naiqiang Tan","orcid":"https://orcid.org/0009-0008-4687-5212"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Naiqiang Tan","raw_affiliation_strings":["Didichuxing Co. Ltd, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Didichuxing Co. Ltd, Beijing, China","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101862104","display_name":"Hui Xiong","orcid":"https://orcid.org/0000-0001-6016-6465"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]},{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Hui Xiong","raw_affiliation_strings":["The Hong Kong University of Science and Technology (Guangzhou), Hong Kong, China and The Hong Kong University of Science and Technology, Hong Kong, China"],"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology (Guangzhou), Hong Kong, China and The Hong Kong University of Science and Technology, Hong Kong, China","institution_ids":["https://openalex.org/I200769079","https://openalex.org/I889458895"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5027644323"],"corresponding_institution_ids":["https://openalex.org/I200769079"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.04029769,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"4969","last_page":"4979"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9969000220298767,"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":0.9969000220298767,"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":0.9940999746322632,"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/T11550","display_name":"Text and Document Classification Technologies","score":0.9853000044822693,"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/computer-science","display_name":"Computer science","score":0.8151096105575562},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.7362416982650757},{"id":"https://openalex.org/keywords/maximization","display_name":"Maximization","score":0.6648967266082764},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.47532275319099426},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.461947500705719},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3998391032218933},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3747060298919678},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.32211875915527344},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.17573857307434082},{"id":"https://openalex.org/keywords/software-engineering","display_name":"Software engineering","score":0.10563638806343079},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.07521432638168335}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8151096105575562},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.7362416982650757},{"id":"https://openalex.org/C2776330181","wikidata":"https://www.wikidata.org/wiki/Q18358244","display_name":"Maximization","level":2,"score":0.6648967266082764},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47532275319099426},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.461947500705719},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3998391032218933},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3747060298919678},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.32211875915527344},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.17573857307434082},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.10563638806343079},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.07521432638168335}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3696410.3714817","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3696410.3714817","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3696410.3714817","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Web Conference 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3696410.3714817","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3696410.3714817","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3696410.3714817","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the ACM on Web Conference 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.5799999833106995}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4409671299.pdf"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W2061820396","https://openalex.org/W2098742124","https://openalex.org/W2606555609","https://openalex.org/W2792990871","https://openalex.org/W2986514296","https://openalex.org/W2990138404","https://openalex.org/W3045462440","https://openalex.org/W3048804154","https://openalex.org/W3081168214","https://openalex.org/W3129831491","https://openalex.org/W3190799236","https://openalex.org/W3197982863","https://openalex.org/W3210450783","https://openalex.org/W3212488946","https://openalex.org/W4307079201","https://openalex.org/W4389524493","https://openalex.org/W4390874070","https://openalex.org/W4396722682","https://openalex.org/W4396722718","https://openalex.org/W4396723256","https://openalex.org/W4396734745","https://openalex.org/W4409657429","https://openalex.org/W6600280466","https://openalex.org/W6600339963"],"related_works":["https://openalex.org/W4205762803","https://openalex.org/W2535856026","https://openalex.org/W2265065644","https://openalex.org/W2947806671","https://openalex.org/W2390878257","https://openalex.org/W2134699697","https://openalex.org/W3017188156","https://openalex.org/W2322875716","https://openalex.org/W2961085424","https://openalex.org/W2144260821"],"abstract_inverted_index":{"Recent":[0],"years":[1],"have":[2],"witnessed":[3],"the":[4,58,117,165,185,216,225],"prevalent":[5],"integration":[6],"of":[7,51,60,148,227],"Large":[8],"Language":[9],"Models":[10],"(LLMs)":[11],"in":[12,231],"various":[13],"Web":[14],"applications,":[15],"such":[16],"as":[17,34,135,144],"search":[18],"engines":[19],"and":[20,62,86,104,139,203],"recommender":[21],"systems.":[22],"As":[23],"an":[24,48,176,206],"emerging":[25],"technique,":[26],"instruction":[27,54,75,121,129,133,186,232],"tuning":[28],"aims":[29],"to":[30,70,88,163,180,193],"align":[31],"pre-trained":[32],"LLMs":[33,61],"capable":[35],"chatbots":[36],"that":[37,46,114,198,215],"excel":[38],"at":[39],"following":[40],"human":[41],"instructions.":[42],"Previous":[43],"research":[44],"indicates":[45],"selecting":[47],"appropriate":[49],"subset":[50,197],"a":[52,110,127,136,159,189,195],"large":[53],"dataset":[55],"can":[56,82,219],"enhance":[57,221],"capabilities":[59],"reduce":[63],"training":[64],"costs.":[65],"However,":[66],"existing":[67],"works":[68],"tend":[69],"overlook":[71],"external":[72],"correlations":[73],"between":[74],"examples":[76],"during":[77],"data":[78,111,229,233],"selection":[79,112],"process,":[80],"which":[81],"introduce":[83],"potential":[84],"bias":[85],"lead":[87],"sub-optimal":[89],"performance.":[90],"To":[91],"bridge":[92],"this":[93,97],"gap,":[94],"we":[95,124,157,174],"formalize":[96],"problem":[98],"from":[99],"graph":[100,137,145,155,161],"influence":[101,107,178,202],"maximization":[102],"perspective":[103],"propose":[105,175],"Uncertainty-aware":[106],"Maximization":[108],"(UniMax),":[109],"framework":[113],"explicitly":[115],"incorporates":[116],"complex":[118],"inter-dependencies":[119],"within":[120],"data.":[122],"Specifically,":[123],"first":[125],"define":[126],"latent":[128,166],"graph,":[130,187],"treating":[131],"each":[132,182],"example":[134,183],"node":[138],"representing":[140],"their":[141],"implicit":[142],"relations":[143],"edges.":[146],"Instead":[147],"solely":[149],"relying":[150],"on":[151,184,211],"heuristic":[152],"metrics":[153],"for":[154],"construction,":[156],"develop":[158],"self-supervised":[160],"learner":[162],"uncover":[164],"structure":[167],"beyond":[168],"surface-level":[169],"feature":[170],"correlations.":[171],"After":[172],"that,":[173],"uncertainty-aware":[177],"function":[179],"score":[181],"allowing":[188],"simple":[190],"greedy":[191],"algorithm":[192],"select":[194],"valuable":[196],"embodies":[199],"both":[200],"high":[201],"uncertainty":[204],"with":[205],"approximation":[207],"guarantee.":[208],"Extensive":[209],"experiments":[210],"public":[212],"datasets":[213],"show":[214],"proposed":[217],"approach":[218],"significantly":[220],"model":[222],"capabilities,":[223],"underscoring":[224],"importance":[226],"exploiting":[228],"dependencies":[230],"selection.":[234]},"counts_by_year":[],"updated_date":"2026-02-26T08:16:20.718346","created_date":"2025-10-10T00:00:00"}