{"id":"https://openalex.org/W7225425909","doi":"https://doi.org/10.5281/zenodo.24010882","title":"Trend-Aware Predictive Generative Inference Caching: The Third Experimental Study on Repetition-Aware Cost Optimization, Independent Quality Evaluation, and Economically Guided Predictive Computation","display_name":"Trend-Aware Predictive Generative Inference Caching: The Third Experimental Study on Repetition-Aware Cost Optimization, Independent Quality Evaluation, and Economically Guided Predictive Computation","publication_year":2026,"publication_date":"2026-11-06","ids":{"openalex":"https://openalex.org/W7225425909","doi":"https://doi.org/10.5281/zenodo.24010882"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.24010882","doi":"https://doi.org/10.5281/zenodo.24010882","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24010882","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.24010882","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100638045","display_name":"Jincheng Zhang","orcid":"https://orcid.org/0000-0001-7332-6704","observed_orcids":["https://orcid.org/0000-0001-7332-6704","https://orcid.org/0009-0005-1860-0009"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Zhang, Jincheng","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0009-0005-1860-0009","affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5100638045"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.21240000426769257,"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":0.21240000426769257,"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":0.19859999418258667,"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/T12127","display_name":"Software System Performance and Reliability","score":0.1371999979019165,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/large-language-models","display_name":"large language models","score":0.7889999747276306},{"id":"https://openalex.org/keywords/demand-forecasting","display_name":"demand forecasting","score":0.7820000052452087},{"id":"https://openalex.org/keywords/computational-cost-reduction","display_name":"computational complexity reduction","score":0.7049999833106995},{"id":"https://openalex.org/keywords/response-quality-evaluation","display_name":"response quality evaluation","score":0.8029999732971191},{"id":"https://openalex.org/keywords/gemma-3","display_name":"Gemma 3","score":0.6990000009536743},{"id":"https://openalex.org/keywords/qwen3","display_name":"Qwen3","score":0.9520000219345093}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7721999883651733},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7213000059127808},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.588100016117096},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.5151000022888184},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5045999884605408},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.4742000102996826},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.45419999957084656},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.4104999899864197},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.4049000144004822},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.39899998903274536},{"id":"https://openalex.org/C134261354","wikidata":"https://www.wikidata.org/wiki/Q938438","display_name":"Statistical inference","level":2,"score":0.3855000138282776},{"id":"https://openalex.org/C917703","wikidata":"https://www.wikidata.org/wiki/Q7239668","display_name":"Predictive inference","level":5,"score":0.3758000135421753},{"id":"https://openalex.org/C90329073","wikidata":"https://www.wikidata.org/wiki/Q914232","display_name":"Ask price","level":2,"score":0.37279999256134033},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.3587999939918518},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.3409999907016754},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.336899995803833},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.33640000224113464},{"id":"https://openalex.org/C87007009","wikidata":"https://www.wikidata.org/wiki/Q210832","display_name":"Statistical hypothesis testing","level":2,"score":0.3303999900817871},{"id":"https://openalex.org/C2776141515","wikidata":"https://www.wikidata.org/wiki/Q1274479","display_name":"Repetition (rhetorical device)","level":2,"score":0.3156999945640564},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.3018999993801117},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.288100004196167},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.28189998865127563},{"id":"https://openalex.org/C2778136018","wikidata":"https://www.wikidata.org/wiki/Q10350689","display_name":"Predictive power","level":2,"score":0.28029999136924744},{"id":"https://openalex.org/C85973986","wikidata":"https://www.wikidata.org/wiki/Q1091731","display_name":"Exploratory research","level":2,"score":0.27799999713897705},{"id":"https://openalex.org/C45804977","wikidata":"https://www.wikidata.org/wiki/Q7239673","display_name":"Predictive modelling","level":2,"score":0.27630001306533813},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.2581999897956848}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.24010882","doi":"https://doi.org/10.5281/zenodo.24010882","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24010882","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},{"id":"doi:10.5281/zenodo.24010883","doi":"https://doi.org/10.5281/zenodo.24010883","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24010883","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.5281/zenodo.24010882","doi":"https://doi.org/10.5281/zenodo.24010882","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24010882","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7225251566","https://openalex.org/W7225332613","https://openalex.org/W7225275234","https://openalex.org/W7225402782","https://openalex.org/W7225398780","https://openalex.org/W7225267314","https://openalex.org/W7225374951","https://openalex.org/W7225427586","https://openalex.org/W7106688525","https://openalex.org/W7225384377"],"abstract_inverted_index":{"The":[0,87,174,208,230,314,400],"increasing":[1],"use":[2],"of":[3,34,71,85,168,191,271,298,308,407],"large":[4],"language":[5],"models":[6],"(LLMs)":[7],"creates":[8],"substantial":[9,32],"computational":[10,69,246],"demand,":[11],"particularly":[12],"when":[13,30],"similar":[14],"or":[15],"identical":[16],"questions":[17,112,120,136,170],"are":[18],"submitted":[19],"repeatedly.":[20],"Conventional":[21],"inference":[22,73],"systems":[23],"generally":[24],"process":[25],"incoming":[26],"requests":[27,36,59,108],"reactively,":[28],"even":[29],"a":[31,129,267,295,331,413],"proportion":[33],"future":[35],"may":[37],"be":[38],"predictable.":[39],"Trend-Aware":[40],"Predictive":[41],"Generative":[42],"Inference":[43],"Caching":[44],"(TPGIC)":[45],"addresses":[46],"this":[47,390],"challenge":[48],"by":[49],"identifying":[50],"potentially":[51],"recurring":[52],"questions,":[53],"preparing":[54],"reusable":[55],"responses":[56,309],"before":[57],"subsequent":[58],"arrive,":[60],"and":[61,98,106,154,196,199,226,259,302,357,411,436],"evaluating":[62],"whether":[63],"predictive":[64,326,437],"preparation":[65],"can":[66,383],"reduce":[67],"the":[68,81,116,164,242,279,305,320,358,379,385,404,408],"cost":[70,228,231,247,269,283,322,398],"repeated":[72],"without":[74],"degrading":[75],"response":[76,127],"quality.":[77],"This":[78,264],"paper":[79],"presents":[80],"third":[82],"experimental":[83,117,376],"study":[84,401],"TPGIC.":[86,263],"experiment":[88,210],"evaluates":[89],"three":[90,105,238,347],"generation":[91,239,336],"models\u2014Gemma":[92],"3":[93,96],"1B,":[94],"Gemma":[95],"4B,":[97],"Qwen3":[99],"4B\u2014under":[100],"two":[101],"repetition":[102],"scenarios":[103],"involving":[104],"ten":[107],"per":[109],"question.":[110],"Twenty":[111],"were":[113,121,177,201],"used":[114,178],"as":[115,179],"benchmark.":[118],"These":[119,373],"generated":[122,304],"using":[123,171,203],"ChatGPT-6":[124,157],"mini":[125,158],"in":[126,310,334,389,417],"to":[128,162,254,266,276,288,294,325,355],"prompt":[130],"requesting":[131],"twenty":[132,192],"specific,":[133],"independently":[134],"answerable":[135],"that":[137,378],"people":[138],"frequently":[139],"ask":[140],"ChatGPT":[141],"across":[142,236,345],"general":[143],"knowledge,":[144],"everyday":[145],"life,":[146],"education,":[147],"writing,":[148],"technology,":[149],"programming,":[150],"health,":[151],"career":[152],"development,":[153],"personal":[155],"productivity.":[156],"was":[159,194],"subsequently":[160],"prompted":[161],"estimate":[163],"monthly":[165],"global":[166],"frequency":[167],"these":[169],"semantic-equivalent":[172],"matching.":[173],"resulting":[175],"estimates":[176],"exploratory":[180],"demand-prediction":[181],"inputs":[182],"rather":[183,328],"than":[184,329],"verified":[185],"usage":[186],"statistics.":[187],"A":[188],"cache":[189],"capacity":[190],"entries":[193],"employed,":[195],"Baseline,":[197],"Demand_Cache,":[198],"TPGIC":[200,274,301],"compared":[202],"an":[204],"independent":[205,216],"evaluator,":[206],"tpgic-qwen3-1.7b.":[207],"completed":[209],"recorded":[211],"2,340":[212],"user":[213],"requests,":[214],"120":[215],"evaluator":[217],"calls,":[218],"360":[219],"quality":[220,339,387],"records,":[221,225],"144":[222],"statistical":[223,361],"test":[224],"18":[227],"records.":[229],"results":[232],"show":[233],"consistent":[234],"patterns":[235],"all":[237,346],"models.":[240],"In":[241,278],"three-repetition":[243],"scenario,":[244,281],"normalized":[245,282],"decreased":[248,284],"from":[249,285,319,330,353],"60":[250],"units":[251,256,261,287,290],"under":[252,257,262,291],"Baseline":[253],"20":[255],"Demand_Cache":[258,303],"6":[260,289],"corresponds":[265],"modeled":[268,296,397],"reduction":[270,297,316,333],"90%":[272],"for":[273,370],"relative":[275],"Baseline.":[277],"ten-repetition":[280],"200":[286],"TPGIC,":[292],"corresponding":[293,312],"97%.":[299],"Importantly,":[300],"same":[306],"number":[307],"each":[311],"scenario.":[313],"additional":[315],"therefore":[317],"arises":[318],"economic":[321,405],"coefficient":[323,410],"assigned":[324],"preparation,":[327],"further":[332],"actual":[335],"calls.":[337],"Independent":[338],"evaluation":[340],"produced":[341],"consistently":[342],"high":[343],"scores":[344,351],"methods.":[348],"Overall":[349],"mean":[350],"ranged":[352],"9.40":[354],"9.70,":[356],"reported":[359],"pairwise":[360],"tests":[362],"did":[363],"not":[364],"identify":[365],"significant":[366],"differences":[367],"after":[368],"adjustment":[369],"multiple":[371],"comparisons.":[372],"findings":[374],"provide":[375],"evidence":[377],"evaluated":[380],"caching":[381],"strategies":[382],"preserve":[384],"observed":[386],"level":[388],"controlled":[391],"repeated-question":[392],"setting":[393],"while":[394],"offering":[395],"different":[396],"profiles.":[399],"also":[402],"clarifies":[403],"interpretation":[406],"0.3":[409],"identifies":[412],"broader":[414],"research":[415],"direction":[416],"which":[418],"future-question":[419],"prediction":[420],"is":[421],"combined":[422],"with":[423],"GPU":[424,434],"workload":[425],"forecasting,":[426],"off-peak":[427],"precomputation,":[428],"dynamic":[429],"electricity-price":[430],"awareness,":[431],"low-cost":[432],"cloud":[433],"selection,":[435],"task":[438],"scheduling.":[439]},"counts_by_year":[],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-11-07T00:00:00"}