{"id":"https://openalex.org/W4403278729","doi":"https://doi.org/10.1109/is61756.2024.10705214","title":"Similarity Thresholds in Retrieval-Augmented Generation","display_name":"Similarity Thresholds in Retrieval-Augmented Generation","publication_year":2024,"publication_date":"2024-08-29","ids":{"openalex":"https://openalex.org/W4403278729","doi":"https://doi.org/10.1109/is61756.2024.10705214"},"language":"en","primary_location":{"id":"doi:10.1109/is61756.2024.10705214","is_oa":false,"landing_page_url":"https://doi.org/10.1109/is61756.2024.10705214","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 12th International Conference on Intelligent Systems (IS)","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/A5022156518","display_name":"Irina Radeva","orcid":"https://orcid.org/0009-0004-8531-5980"},"institutions":[{"id":"https://openalex.org/I24768866","display_name":"Bulgarian Academy of Sciences","ror":"https://ror.org/01x8hew03","country_code":"BG","type":"government","lineage":["https://openalex.org/I24768866"]},{"id":"https://openalex.org/I4210163063","display_name":"Institute of Information and Communication Technologies","ror":"https://ror.org/05fpsjc82","country_code":"BG","type":"facility","lineage":["https://openalex.org/I24768866","https://openalex.org/I4210163063"]}],"countries":["BG"],"is_corresponding":false,"raw_author_name":"Irina Radeva","raw_affiliation_strings":["Institute of Information and Communication Technologies – Bulgarian Academy of Sciences,Intelligent Systems Dept.,Sofia,Bulgaria"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Information and Communication Technologies – Bulgarian Academy of Sciences,Intelligent Systems Dept.,Sofia,Bulgaria","institution_ids":["https://openalex.org/I4210163063","https://openalex.org/I24768866"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037805365","display_name":"\u0418\u0432\u0430\u043d \u041f\u043e\u043f\u0447\u0435\u0432","orcid":"https://orcid.org/0009-0002-8128-3980"},"institutions":[{"id":"https://openalex.org/I24768866","display_name":"Bulgarian Academy of Sciences","ror":"https://ror.org/01x8hew03","country_code":"BG","type":"government","lineage":["https://openalex.org/I24768866"]}],"countries":["BG"],"is_corresponding":false,"raw_author_name":"Ivan Popchev","raw_affiliation_strings":["Bulgarian Academy of Sciences,Sofia,Bulgaria"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Bulgarian Academy of Sciences,Sofia,Bulgaria","institution_ids":["https://openalex.org/I24768866"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5083327971","display_name":"Miroslava Dimitrova","orcid":"https://orcid.org/0009-0008-7034-9773"},"institutions":[{"id":"https://openalex.org/I24768866","display_name":"Bulgarian Academy of Sciences","ror":"https://ror.org/01x8hew03","country_code":"BG","type":"government","lineage":["https://openalex.org/I24768866"]},{"id":"https://openalex.org/I4210163063","display_name":"Institute of Information and Communication Technologies","ror":"https://ror.org/05fpsjc82","country_code":"BG","type":"facility","lineage":["https://openalex.org/I24768866","https://openalex.org/I4210163063"]}],"countries":["BG"],"is_corresponding":false,"raw_author_name":"Miroslava Dimitrova","raw_affiliation_strings":["Institute of Information and Communication Technologies – Bulgarian Academy of Sciences,Intelligent Systems Dept.,Sofia,Bulgaria"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Information and Communication Technologies – Bulgarian Academy of Sciences,Intelligent Systems Dept.,Sofia,Bulgaria","institution_ids":["https://openalex.org/I4210163063","https://openalex.org/I24768866"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.4437,"has_fulltext":false,"cited_by_count":8,"citation_normalized_percentile":{"value":0.9061747,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9072999954223633,"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":0.9072999954223633,"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":0.6725499629974365},{"id":"https://openalex.org/keywords/similarity","display_name":"Similarity (geometry)","score":0.6498488187789917},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5604029893875122},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.4410310983657837},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3774866461753845},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.07172581553459167}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6725499629974365},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.6498488187789917},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5604029893875122},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.4410310983657837},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3774866461753845},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.07172581553459167}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/is61756.2024.10705214","is_oa":false,"landing_page_url":"https://doi.org/10.1109/is61756.2024.10705214","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 12th International Conference on Intelligent Systems (IS)","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":14,"referenced_works":["https://openalex.org/W172260869","https://openalex.org/W1846261984","https://openalex.org/W2296437741","https://openalex.org/W3099700870","https://openalex.org/W4241153016","https://openalex.org/W4310175361","https://openalex.org/W4318002288","https://openalex.org/W4389519118","https://openalex.org/W4389524176","https://openalex.org/W4393948917","https://openalex.org/W4399530612","https://openalex.org/W6678262379","https://openalex.org/W6682631176","https://openalex.org/W6898505805"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052","https://openalex.org/W2033914206","https://openalex.org/W2042327336"],"abstract_inverted_index":{"This":[0,74],"paper":[1],"evaluates":[2],"the":[3,17,26,31,37,46,77,82,89,97,119,141,176,179,184,211,215],"performance":[4,39,143,159],"of":[5,19,25,91,108,183,214],"open-source":[6],"Large":[7],"Language":[8,42],"Models":[9],"(LLMs)":[10],"Mistral:":[11],"7b,":[12],"Llama2:7b,":[13],"and":[14,71,149,166,204],"Orca2:7b":[15],"within":[16],"context":[18],"Retrieval-Augmented":[20],"Generation":[21],"(RAG).":[22],"The":[23,111,127,168],"purpose":[24],"study":[27,75,169],"is":[28,114],"to":[29,52,67,104,118,202],"identify":[30],"similarity":[32,78,136,180],"score":[33,79,181],"thresholds":[34,80,162,182],"that":[35,130,173],"yield":[36],"best":[38,133],"across":[40],"Natural":[41],"Processing":[43],"metrics.":[44],"Despite":[45],"advancements":[47],"in":[48,178],"LLMs,":[49],"their":[50],"ability":[51],"access":[53],"up-to-date,":[54],"domain-specific":[55],"information":[56],"remains":[57],"limited.":[58],"RAG":[59],"addresses":[60],"this":[61],"by":[62],"integrating":[63],"external":[64],"database":[65],"querying":[66],"produce":[68],"contextually":[69],"comprehensive":[70],"precise":[72],"responses.":[73],"investigates":[76],"for":[81],"LLMs":[83,197],"using":[84,96],"a":[85,115,145,152],"custom-developed":[86],"dataset":[87],"from":[88],"domain":[90],"agriculture.":[92],"Tests":[93],"were":[94],"conducted":[95],"PaSSER":[98,112,208],"web-based":[99],"application,":[100],"which":[101],"was":[102],"enhanced":[103],"support":[105],"detailed":[106],"configuration":[107],"retrieval":[109],"processes.":[110],"App":[113,209],"complementary":[116],"project":[117],"Smart":[120],"Crop":[121],"Production":[122],"Data":[123],"Exchange":[124],"(SCPDx)":[125],"platform.":[126,217],"results":[128],"showed":[129],"Mistral":[131],"performs":[132],"at":[134,144,151,160],"lower":[135,153],"threshold":[137,147,154],"(0.55),":[138,155],"Orca":[139],"achieves":[140],"highest":[142],"moderate":[146],"(0.65),":[148],"Llama":[150],"but":[156],"has":[157],"strong":[158],"higher":[161],"as":[163],"well":[164],"(0.70":[165],"0.85).":[167],"discusses":[170],"several":[171],"factors":[172],"could":[174],"cause":[175],"differences":[177],"assessed":[185],"LLMs.":[186],"Future":[187],"development":[188],"will":[189],"focus":[190],"on":[191,207],"leveraging":[192],"other":[193],"pre-trained":[194],"open":[195],"source":[196],"(over":[198],"40b),":[199],"exploring":[200],"approaches":[201],"fine-tuning,":[203],"further":[205],"integration":[206],"with":[210],"existing":[212],"infrastructure":[213],"SCPDx":[216]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":6}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}