{"id":"https://openalex.org/W7167508910","doi":"https://doi.org/10.13052/jwe1540-9589.2552","title":"Hybrid Layered Retrieval and Task-aware Embeddings for Efficient Persian RAG Systems","display_name":"Hybrid Layered Retrieval and Task-aware Embeddings for Efficient Persian RAG Systems","publication_year":2026,"publication_date":"2026-07-06","ids":{"openalex":"https://openalex.org/W7167508910","doi":"https://doi.org/10.13052/jwe1540-9589.2552"},"language":null,"primary_location":{"id":"doi:10.13052/jwe1540-9589.2552","is_oa":true,"landing_page_url":"https://doi.org/10.13052/jwe1540-9589.2552","pdf_url":null,"source":{"id":"https://openalex.org/S114499935","display_name":"Journal of Web Engineering","issn_l":"1540-9589","issn":["1540-9589","1544-5976"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4366111290","host_organization_name":"River Publishers","host_organization_lineage":["https://openalex.org/P4366111290"],"host_organization_lineage_names":["River Publishers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Web Engineering","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.13052/jwe1540-9589.2552","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069384181","display_name":"Toktam Zoughi","orcid":"https://orcid.org/0000-0002-1797-6910"},"institutions":[{"id":"https://openalex.org/I4210089766","display_name":"Technical and Vocational University","ror":"https://ror.org/00854zy02","country_code":"IR","type":"education","lineage":["https://openalex.org/I4210089766"]}],"countries":["IR"],"is_corresponding":false,"raw_author_name":"Toktam Zoughi","raw_affiliation_strings":["Department of Computer Engineering, Technical and Vocational University (TVU), Tehran, Iran"],"raw_orcid":"https://orcid.org/0000-0002-1797-6910","affiliations":[{"raw_affiliation_string":"Department of Computer Engineering, Technical and Vocational University (TVU), Tehran, Iran","institution_ids":["https://openalex.org/I4210089766"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003017527","display_name":"Ehsan Arianyan","orcid":"https://orcid.org/0000-0001-8215-7797"},"institutions":[{"id":"https://openalex.org/I4210102178","display_name":"ICT Research Institute","ror":"https://ror.org/01a3g2z22","country_code":"IR","type":"facility","lineage":["https://openalex.org/I4210102178"]}],"countries":["IR"],"is_corresponding":false,"raw_author_name":"Ehsan Arianyan","raw_affiliation_strings":["Information Technology Faculty, ICT Research Institute, Tehran, Iran"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Information Technology Faculty, ICT Research Institute, Tehran, Iran","institution_ids":["https://openalex.org/I4210102178"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100915147","display_name":"Maryam Mahmoudi","orcid":null},"institutions":[{"id":"https://openalex.org/I4210102178","display_name":"ICT Research Institute","ror":"https://ror.org/01a3g2z22","country_code":"IR","type":"facility","lineage":["https://openalex.org/I4210102178"]}],"countries":["IR"],"is_corresponding":false,"raw_author_name":"Maryam Mahmoudi","raw_affiliation_strings":["Information Technology Faculty, ICT Research Institute, Tehran, Iran"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Information Technology Faculty, ICT Research Institute, Tehran, Iran","institution_ids":["https://openalex.org/I4210102178"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140094592","display_name":"Mahtab Aghdamifard","orcid":null},"institutions":[{"id":"https://openalex.org/I4210089766","display_name":"Technical and Vocational University","ror":"https://ror.org/00854zy02","country_code":"IR","type":"education","lineage":["https://openalex.org/I4210089766"]}],"countries":["IR"],"is_corresponding":false,"raw_author_name":"Mahtab Aghdamifard","raw_affiliation_strings":["Department of Computer Engineering, Technical and Vocational University (TVU), Tehran, Iran"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Engineering, Technical and Vocational University (TVU), Tehran, Iran","institution_ids":["https://openalex.org/I4210089766"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5140148179","display_name":"Mohadese Nikoogoftar","orcid":null},"institutions":[{"id":"https://openalex.org/I4210089766","display_name":"Technical and Vocational University","ror":"https://ror.org/00854zy02","country_code":"IR","type":"education","lineage":["https://openalex.org/I4210089766"]}],"countries":["IR"],"is_corresponding":false,"raw_author_name":"Mohadese Nikoogoftar","raw_affiliation_strings":["Department of Computer Engineering, Technical and Vocational University (TVU), Tehran, Iran"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Engineering, Technical and Vocational University (TVU), Tehran, Iran","institution_ids":["https://openalex.org/I4210089766"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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":0.91493105,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"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":0.6211000084877014,"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":0.6211000084877014,"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":0.1467999964952469,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.13359999656677246,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/language-model","display_name":"Language model","score":0.4311999976634979},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.3873000144958496},{"id":"https://openalex.org/keywords/word-embedding","display_name":"Word embedding","score":0.3779999911785126},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.35670000314712524},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.3521000146865845},{"id":"https://openalex.org/keywords/language-identification","display_name":"Language identification","score":0.3452000021934509},{"id":"https://openalex.org/keywords/similarity","display_name":"Similarity (geometry)","score":0.32109999656677246},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.3050999939441681}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8413000106811523},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5192999839782715},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5065000057220459},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4311999976634979},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.423799991607666},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.3873000144958496},{"id":"https://openalex.org/C2777462759","wikidata":"https://www.wikidata.org/wiki/Q18395344","display_name":"Word embedding","level":3,"score":0.3779999911785126},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.35670000314712524},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.3521000146865845},{"id":"https://openalex.org/C129792486","wikidata":"https://www.wikidata.org/wiki/Q1050419","display_name":"Language identification","level":3,"score":0.3452000021934509},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.32109999656677246},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.3050999939441681},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.30160000920295715},{"id":"https://openalex.org/C50897621","wikidata":"https://www.wikidata.org/wiki/Q2665508","display_name":"Hybrid system","level":2,"score":0.29989999532699585},{"id":"https://openalex.org/C183115368","wikidata":"https://www.wikidata.org/wiki/Q856577","display_name":"Weighting","level":2,"score":0.2754000127315521},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.26750001311302185},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.26660001277923584},{"id":"https://openalex.org/C197947376","wikidata":"https://www.wikidata.org/wiki/Q5155608","display_name":"Comparability","level":2,"score":0.2660999894142151},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.25609999895095825},{"id":"https://openalex.org/C161156560","wikidata":"https://www.wikidata.org/wiki/Q1638872","display_name":"Document retrieval","level":2,"score":0.25609999895095825},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2524999976158142},{"id":"https://openalex.org/C500882744","wikidata":"https://www.wikidata.org/wiki/Q269236","display_name":"Latent Dirichlet allocation","level":3,"score":0.2524000108242035},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.13052/jwe1540-9589.2552","is_oa":true,"landing_page_url":"https://doi.org/10.13052/jwe1540-9589.2552","pdf_url":null,"source":{"id":"https://openalex.org/S114499935","display_name":"Journal of Web Engineering","issn_l":"1540-9589","issn":["1540-9589","1544-5976"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4366111290","host_organization_name":"River Publishers","host_organization_lineage":["https://openalex.org/P4366111290"],"host_organization_lineage_names":["River Publishers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Web Engineering","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.13052/jwe1540-9589.2552","is_oa":true,"landing_page_url":"https://doi.org/10.13052/jwe1540-9589.2552","pdf_url":null,"source":{"id":"https://openalex.org/S114499935","display_name":"Journal of Web Engineering","issn_l":"1540-9589","issn":["1540-9589","1544-5976"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4366111290","host_organization_name":"River Publishers","host_organization_lineage":["https://openalex.org/P4366111290"],"host_organization_lineage_names":["River Publishers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Journal of Web Engineering","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"This":[0,111],"study":[1],"introduces":[2],"task-injected":[3,208],"layered":[4,35,76,209],"hybrid":[5,40,88,210],"retrieval-augmented":[6,211,225],"generation":[7,180,212,226],"(TILHR-RAG),":[8],"a":[9,34,39,46,101,106,214],"framework":[10],"specifically":[11],"designed":[12],"for":[13,61,182,220,227,242],"Persian":[14,183,221],"to":[15],"address":[16],"the":[17,23,129],"scarcity":[18],"of":[19,25],"native-language":[20],"resources":[21],"and":[22,38,56,100,116,135,144,179,187,198,217],"limitations":[24],"English-centric":[26],"approaches.":[27],"The":[28,75],"architecture":[29],"combines":[30],"task-aware":[31],"query":[32],"augmentation,":[33],"retrieval":[36,66,178],"strategy,":[37],"semantic\u2013lexical":[41],"retriever,":[42],"all":[43],"supported":[44],"by":[45,105,191],"multi-stage":[47],"pipeline":[48],"that":[49,126,160],"includes":[50],"preprocessing,":[51],"document":[52,89],"chunking,":[53],"question":[54,222],"generation,":[55],"embedding.":[57],"A":[58],"novel":[59],"mechanism":[60],"injecting":[62],"task-specific":[63],"vectors":[64],"directs":[65],"toward":[67],"domain":[68],"intent":[69],"while":[70,237],"preserving":[71],"comparability":[72],"across":[73,120],"documents.":[74],"design":[77],"operates":[78],"in":[79],"three":[80],"stages:":[81],"per-task":[82],"frequently":[83],"asked":[84],"questions":[85],"(FAQ)":[86],"retrieval,":[87],"search":[90],"using":[91],"FAISS":[92],"semantic":[93,139],"similarity":[94],"combined":[95],"with":[96,141],"BM25":[97],"keyword":[98],"matching,":[99],"fallback":[102],"response":[103],"generated":[104],"large":[107,233],"language":[108,234],"model":[109,235],"(LLM).":[110],"structure":[112],"ensures":[113],"both":[114],"precision":[115],"robustness.":[117],"Comprehensive":[118],"experiments":[119],"five":[121],"progressively":[122],"refined":[123],"configurations":[124],"demonstrate":[125],"TILHR-RAG":[127],"achieves":[128],"best":[130],"balance":[131],"among":[132],"accuracy,":[133],"efficiency,":[134],"scalability,":[136],"reaching":[137],"89.67%":[138],"accuracy":[140,161],"moderate":[142],"latency":[143,168],"memory":[145],"consumption":[146],"on":[147,153],"NVIDIA":[148],"A100":[149],"hardware.":[150],"Further":[151],"evaluations":[152],"low-resource":[154,228],"graphics":[155],"processing":[156],"units":[157],"(GPUs)":[158],"confirm":[159],"remains":[162],"stable":[163],"under":[164],"hardware":[165],"constraints,":[166],"although":[167],"increases":[169],"significantly.":[170],"Moreover,":[171],"multilingual":[172,244],"E5":[173],"embedding":[174],"models":[175],"substantially":[176],"improve":[177],"quality":[181],"\u2013":[184,190],"outperforming":[185],"ParsBERT":[186],"Sentence-BERT":[188],"(SBERT)":[189],"mitigating":[192],"challenges":[193],"such":[194],"as":[195,213],"orthographic":[196],"variation":[197],"complex":[199],"compound":[200],"word":[201],"structures.":[202],"Taken":[203],"together,":[204],"these":[205],"findings":[206],"establish":[207],"practical,":[215],"reproducible,":[216],"resource-efficient":[218],"blueprint":[219],"answering,":[223],"advancing":[224],"languages":[229],"without":[230],"requiring":[231],"costly":[232],"fine-tuning,":[236],"also":[238],"offering":[239],"adaptable":[240],"strategies":[241],"broader":[243],"applications.":[245]},"counts_by_year":[],"updated_date":"2026-07-08T06:17:01.165560","created_date":"2026-07-07T00:00:00"}