{"id":"https://openalex.org/W7125959561","doi":"https://doi.org/10.1109/smc58881.2025.11343731","title":"KiC: Keyword-inspired Cascade for Cost-Efficient Text Generation with LLMs","display_name":"KiC: Keyword-inspired Cascade for Cost-Efficient Text Generation with LLMs","publication_year":2025,"publication_date":"2025-10-05","ids":{"openalex":"https://openalex.org/W7125959561","doi":"https://doi.org/10.1109/smc58881.2025.11343731"},"language":null,"primary_location":{"id":"doi:10.1109/smc58881.2025.11343731","is_oa":false,"landing_page_url":"https://doi.org/10.1109/smc58881.2025.11343731","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Systems, Man, and Cybernetics (SMC)","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/A5101604428","display_name":"Woo-Chan Kim","orcid":"https://orcid.org/0000-0002-5373-5508"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Woo-Chan Kim","raw_affiliation_strings":["Korea University,Department of Artificial Intelligence,Seoul,Korea,02841"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Department of Artificial Intelligence,Seoul,Korea,02841","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111978986","display_name":"Ji-Hoon Park","orcid":null},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Ji-Hoon Park","raw_affiliation_strings":["Korea University,Department of Artificial Intelligence,Seoul,Korea,02841"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Department of Artificial Intelligence,Seoul,Korea,02841","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5124139481","display_name":"Seong-Whan Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seong-Whan Lee","raw_affiliation_strings":["Korea University,Department of Artificial Intelligence,Seoul,Korea,02841"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Korea University,Department of Artificial Intelligence,Seoul,Korea,02841","institution_ids":["https://openalex.org/I197347611"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"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":0.80494811,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"6365","last_page":"6370"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.5936999917030334,"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/T10028","display_name":"Topic Modeling","score":0.5936999917030334,"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":0.08020000159740448,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.032499998807907104,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/cascade","display_name":"Cascade","score":0.802299976348877},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7045999765396118},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.5267999768257141},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4943999946117401},{"id":"https://openalex.org/keywords/natural-language-generation","display_name":"Natural language generation","score":0.48240000009536743},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.4361000061035156},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.40700000524520874},{"id":"https://openalex.org/keywords/production","display_name":"Production (economics)","score":0.34700000286102295}],"concepts":[{"id":"https://openalex.org/C34146451","wikidata":"https://www.wikidata.org/wiki/Q5048094","display_name":"Cascade","level":2,"score":0.802299976348877},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7045999765396118},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6901999711990356},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.5267999768257141},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4943999946117401},{"id":"https://openalex.org/C2776187449","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Natural language generation","level":3,"score":0.48240000009536743},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4666000008583069},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4620000123977661},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.4361000061035156},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.40700000524520874},{"id":"https://openalex.org/C2778348673","wikidata":"https://www.wikidata.org/wiki/Q739302","display_name":"Production (economics)","level":2,"score":0.34700000286102295},{"id":"https://openalex.org/C2985684807","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Text generation","level":2,"score":0.34540000557899475},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.34060001373291016},{"id":"https://openalex.org/C3018949938","wikidata":"https://www.wikidata.org/wiki/Q17166101","display_name":"Text messaging","level":2,"score":0.2896000146865845},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.2827000021934509},{"id":"https://openalex.org/C155092808","wikidata":"https://www.wikidata.org/wiki/Q182557","display_name":"Computational linguistics","level":2,"score":0.28139999508857727},{"id":"https://openalex.org/C2776608160","wikidata":"https://www.wikidata.org/wiki/Q4785462","display_name":"Natural (archaeology)","level":2,"score":0.27160000801086426},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2623000144958496},{"id":"https://openalex.org/C90559484","wikidata":"https://www.wikidata.org/wiki/Q778379","display_name":"Expression (computer science)","level":2,"score":0.26109999418258667}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/smc58881.2025.11343731","is_oa":false,"landing_page_url":"https://doi.org/10.1109/smc58881.2025.11343731","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Systems, Man, and Cybernetics (SMC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.7685177326202393,"display_name":"Quality Education"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320321373","display_name":"Korea University","ror":"https://ror.org/047dqcg40"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":12,"referenced_works":["https://openalex.org/W2011013898","https://openalex.org/W2061247098","https://openalex.org/W2104558951","https://openalex.org/W2170682101","https://openalex.org/W2970641574","https://openalex.org/W3007994488","https://openalex.org/W3035104236","https://openalex.org/W3201174429","https://openalex.org/W4294170691","https://openalex.org/W4311887664","https://openalex.org/W4385573392","https://openalex.org/W4385825586"],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1,14],"models":[2,19],"(LLMs)":[3],"have":[4],"demonstrated":[5],"state-of-the-art":[6],"performance":[7],"across":[8],"a":[9,37,43,56,84,102,133,164],"wide":[10],"range":[11],"of":[12,65,110,119,148],"natural":[13],"processing":[15],"tasks.":[16],"However,":[17],"high-performing":[18],"are":[20],"typically":[21],"accessible":[22],"only":[23,46],"via":[24],"APIs,":[25],"incurring":[26],"substantial":[27],"inference":[28],"costs.":[29],"Cascade":[30,82],"methods":[31],"address":[32],"this":[33],"by":[34,155],"initially":[35],"employing":[36],"cheaper":[38],"model":[39,104],"and":[40,60,105,159],"escalating":[41],"to":[42,54,124,132],"stronger":[44,134],"one":[45],"when":[47],"necessary.":[48],"Nevertheless,":[49],"existing":[50],"cascade":[51],"approaches":[52],"struggle":[53],"select":[55],"reliable":[57],"representative":[58,96],"response":[59],"assess":[61],"the":[62,94,107,117,126],"overall":[63],"reliability":[64],"free-form":[66,89,139],"outputs,":[67],"as":[68],"they":[69],"rely":[70],"on":[71,116,137,157],"exact":[72],"text":[73,90,140],"matching.":[74],"To":[75],"overcome":[76],"these":[77],"limitations,":[78],"we":[79],"propose":[80],"Keyword-inspired":[81],"(KiC),":[83],"novel":[85],"framework":[86],"for":[87],"cost-efficient":[88],"generation.":[91],"KiC":[92,121,145],"identifies":[93],"most":[95],"answer":[97],"among":[98],"multiple":[99],"outputs":[100],"from":[101],"weaker":[103,127],"evaluates":[106],"semantic":[108],"alignment":[109],"other":[111],"responses":[112],"with":[113],"it.":[114],"Based":[115],"degree":[118],"alignment,":[120],"determines":[122],"whether":[123],"accept":[125],"model\u2019s":[128],"output":[129],"or":[130],"escalate":[131],"model.":[135],"Experiments":[136],"three":[138],"generation":[141],"benchmarks":[142],"show":[143],"that":[144],"achieves":[146],"97.53%":[147],"GPT-4\u2019s":[149],"accuracy":[150],"while":[151],"reducing":[152],"API":[153],"costs":[154],"28.81%":[156],"average,":[158],"even":[160],"outperforms":[161],"GPT-4":[162],"in":[163],"specific":[165],"benchmark.":[166]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-01-29T00:00:00"}