{"id":"https://openalex.org/W7225759770","doi":"https://doi.org/10.5281/zenodo.24097814","title":"Beyond Vector Similarity: A Two-Stage Safety-Aware Semantic Caching Architecture for Large Language Model Gateways","display_name":"Beyond Vector Similarity: A Two-Stage Safety-Aware Semantic Caching Architecture for Large Language Model Gateways","publication_year":2026,"publication_date":"2026-11-10","ids":{"openalex":"https://openalex.org/W7225759770","doi":"https://doi.org/10.5281/zenodo.24097814"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.24097814","doi":"https://doi.org/10.5281/zenodo.24097814","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24097814","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.24097814","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5161386109","display_name":"MappillaiMeeran A","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"A, MappillaiMeeran","raw_affiliation_strings":["Sri Manakula Vinayagar Engineering College"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sri Manakula Vinayagar Engineering College","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5161386109"],"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/T10028","display_name":"Topic Modeling","score":0.7896000146865845,"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":0.7896000146865845,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.08460000157356262,"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.05249999836087227,"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"}}],"keywords":[],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8345999717712402},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.6603999733924866},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.4268999993801117},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4025999903678894},{"id":"https://openalex.org/C2780762811","wikidata":"https://www.wikidata.org/wiki/Q1784941","display_name":"Cosine similarity","level":3,"score":0.3828999996185303},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.3711000084877014},{"id":"https://openalex.org/C130318100","wikidata":"https://www.wikidata.org/wiki/Q2268914","display_name":"Semantic similarity","level":2,"score":0.36899998784065247},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.36480000615119934},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35749998688697815},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.35569998621940613},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.3269999921321869},{"id":"https://openalex.org/C37926939","wikidata":"https://www.wikidata.org/wiki/Q7449061","display_name":"Semantic equivalence","level":4,"score":0.3172000050544739},{"id":"https://openalex.org/C167713795","wikidata":"https://www.wikidata.org/wiki/Q352090","display_name":"Smart Cache","level":5,"score":0.3158000111579895},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.31470000743865967},{"id":"https://openalex.org/C45874996","wikidata":"https://www.wikidata.org/wiki/Q37045","display_name":"Markup language","level":3,"score":0.296099990606308},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2840999960899353},{"id":"https://openalex.org/C2780069185","wikidata":"https://www.wikidata.org/wiki/Q7977945","display_name":"Equivalence (formal languages)","level":2,"score":0.2750999927520752},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.26499998569488525}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.24097814","doi":"https://doi.org/10.5281/zenodo.24097814","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24097814","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.24097815","doi":"https://doi.org/10.5281/zenodo.24097815","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24097815","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.24097814","doi":"https://doi.org/10.5281/zenodo.24097814","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24097814","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":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Model":[2],"(LLM)":[3],"gateways":[4],"rely":[5],"heavily":[6],"on":[7,32,151],"semantic":[8,27,79,140],"caching":[9,80],"to":[10,40,49,71,99],"reduce":[11],"inference":[12],"latency":[13,160],"and":[14,58,115,125,153],"computational":[15],"cost":[16],"by":[17],"reusing":[18],"responses":[19,68],"for":[20,83],"semantically":[21],"similar":[22],"queries.":[23],"However,":[24],"conventional":[25],"single-stage":[26],"caches":[28],"that":[29,119],"depend":[30],"solely":[31],"dense":[33,95],"vector":[34,45,96],"cosine":[35],"similarity":[36,46,107],"are":[37],"structurally":[38],"vulnerable":[39],"false-positive":[41,148],"cache":[42,102,131,149],"hits.":[43],"High":[44],"frequently":[47],"fails":[48],"distinguish":[50],"logical":[51],"negations,":[52],"antonym":[53],"substitutions,":[54],"subtle":[55],"parameter":[56],"shifts,":[57],"safety-critical":[59],"boundary":[60],"conditions\u2014resulting":[61],"in":[62],"contextually":[63],"incorrect":[64],"or":[65],"unsafe":[66],"cached":[67],"being":[69],"served":[70],"users.":[72],"This":[73],"paper":[74],"presents":[75],"a":[76,112,130],"two-stage":[77],"safety-aware":[78],"architecture":[81,145],"designed":[82],"production":[84],"LLM":[85],"gateways.":[86],"Stage":[87,109],"1":[88],"performs":[89],"fast":[90,135],"approximate":[91],"nearest-neighbor":[92],"retrieval":[93,137],"using":[94],"embeddings":[97],"(pgvector)":[98],"identify":[100],"candidate":[101,136],"entries":[103],"above":[104],"an":[105],"initial":[106],"threshold.":[108],"2":[110],"introduces":[111],"deterministic":[113],"safety":[114],"intent":[116],"verification":[117],"gate":[118],"evaluates":[120],"structural":[121],"divergence,":[122],"negation":[123],"tokens,":[124],"policy":[126],"boundaries":[127],"before":[128],"confirming":[129],"hit.":[132],"By":[133],"decoupling":[134],"from":[138],"strict":[139],"equivalence":[141],"verification,":[142],"the":[143],"proposed":[144],"significantly":[146],"reduces":[147],"hits":[150],"adversarial":[152],"edge-case":[154],"prompts":[155],"while":[156],"maintaining":[157],"low":[158],"end-to-end":[159],"overhead.":[161]},"counts_by_year":[],"updated_date":"2026-11-11T12:16:26.678958","created_date":"2026-11-11T00:00:00"}