{"id":"https://openalex.org/W7119709855","doi":"https://doi.org/10.36227/techrxiv.176800904.47969417/v1","title":"HordeVision: An Open-Source Kazakh Vision-Language Model","display_name":"HordeVision: An Open-Source Kazakh Vision-Language Model","publication_year":2026,"publication_date":"2026-01-10","ids":{"openalex":"https://openalex.org/W7119709855","doi":"https://doi.org/10.36227/techrxiv.176800904.47969417/v1"},"language":"en","primary_location":{"id":"doi:10.36227/techrxiv.176800904.47969417/v1","doi":"https://doi.org/10.36227/techrxiv.176800904.47969417/v1","is_oa":true,"landing_page_url":"https://doi.org/10.36227/techrxiv.176800904.47969417/v1","pdf_url":"https://www.techrxiv.org/doi/pdf/10.36227/techrxiv.176800904.47969417/v1","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"posted-content"},"volume":null,"type":"preprint","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://www.techrxiv.org/doi/pdf/10.36227/techrxiv.176800904.47969417/v1","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5122663974","display_name":"Pavel Zubitskii","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pavel Zubitskii","raw_affiliation_strings":["Horde Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Horde Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122539682","display_name":"Vitaliy Morozov","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vitaliy Morozov","raw_affiliation_strings":["Horde Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Horde Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031772577","display_name":"Sanzhar Murzakhmetov","orcid":"https://orcid.org/0000-0001-6494-8982","observed_orcids":["https://orcid.org/0000-0001-6494-8982"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sanzhar Murzakhmetov","raw_affiliation_strings":["Horde Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Horde Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119819314","display_name":"Beksultan Sagyndyk","orcid":"https://orcid.org/0009-0003-6957-2680","observed_orcids":["https://orcid.org/0009-0003-6957-2680"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Beksultan Sagyndyk","raw_affiliation_strings":["Horde Research"],"raw_orcid":"https://orcid.org/0009-0003-6957-2680","affiliations":[{"raw_affiliation_string":"Horde Research","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5098921390","display_name":"Sanzhar Umbet","orcid":"https://orcid.org/0000-0003-0506-7450","observed_orcids":["https://orcid.org/0000-0003-0506-7450"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sanzhar Umbet","raw_affiliation_strings":["Horde Research"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Horde Research","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9994999766349792,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9994999766349792,"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"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":9.999999747378752e-05,"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/T10601","display_name":"Handwritten Text Recognition Techniques","score":0.0,"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/kazakh-language","display_name":"Kazakh language","score":0.9409999847412109},{"id":"https://openalex.org/keywords/vision-language-models","display_name":"vision-language models","score":0.9599999785423279},{"id":"https://openalex.org/keywords/optical-character-recognition","display_name":"optical character recognition","score":0.8119999766349792},{"id":"https://openalex.org/keywords/image-captioning","display_name":"image captioning","score":0.9990000128746033},{"id":"https://openalex.org/keywords/visual-question-answering","display_name":"visual question answering","score":1.0},{"id":"https://openalex.org/keywords/multimodal-artificial-intelligence","display_name":"multimodal AI","score":0.7009999752044678},{"id":"https://openalex.org/keywords/synthetic-data-generation","display_name":"synthetic data generation","score":0.5429999828338623},{"id":"https://openalex.org/keywords/group-relative-policy-optimization","display_name":"Group Relative Policy Optimization","score":0.8119999766349792},{"id":"https://openalex.org/keywords/reinforcement-learning-from-human-feedback","display_name":"reinforcement learning from human feedback","score":0.9670000076293945},{"id":"https://openalex.org/keywords/supervised-fine-tuning","display_name":"supervised fine-tuning","score":0.8560000061988831},{"id":"https://openalex.org/keywords/low-resource-languages","display_name":"low-resource languages","score":0.6110000014305115},{"id":"https://openalex.org/keywords/open-source-ai","display_name":"open-source AI","score":0.7279999852180481}],"concepts":[{"id":"https://openalex.org/C2781297163","wikidata":"https://www.wikidata.org/wiki/Q9252","display_name":"Kazakh","level":2,"score":0.854200005531311},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5455999970436096},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.44780001044273376},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.44690001010894775},{"id":"https://openalex.org/C2780861071","wikidata":"https://www.wikidata.org/wiki/Q1062934","display_name":"Character (mathematics)","level":2,"score":0.4162999987602234},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.39959999918937683},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.33399999141693115},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.31369999051094055},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.28679999709129333},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.2651999890804291}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.36227/techrxiv.176800904.47969417/v1","doi":"https://doi.org/10.36227/techrxiv.176800904.47969417/v1","is_oa":true,"landing_page_url":"https://doi.org/10.36227/techrxiv.176800904.47969417/v1","pdf_url":"https://www.techrxiv.org/doi/pdf/10.36227/techrxiv.176800904.47969417/v1","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"posted-content"}],"best_oa_location":{"id":"doi:10.36227/techrxiv.176800904.47969417/v1","doi":"https://doi.org/10.36227/techrxiv.176800904.47969417/v1","is_oa":true,"landing_page_url":"https://doi.org/10.36227/techrxiv.176800904.47969417/v1","pdf_url":"https://www.techrxiv.org/doi/pdf/10.36227/techrxiv.176800904.47969417/v1","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"posted-content"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7119709855.pdf","grobid_xml":"https://content.openalex.org/works/W7119709855.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7140114651","https://openalex.org/W4405030061","https://openalex.org/W7197673002","https://openalex.org/W4417052556","https://openalex.org/W7140090158","https://openalex.org/W7213885098","https://openalex.org/W6929653720","https://openalex.org/W6903759483","https://openalex.org/W4386185600","https://openalex.org/W7169538217"],"abstract_inverted_index":{"We":[0,62],"present":[1],"HordeVision,":[2],"an":[3],"open-source":[4,26],"Kazakh":[5],"vision-language":[6],"model":[7],"designed":[8],"for":[9,33],"optical":[10],"character":[11],"recognition":[12],"(OCR),":[13],"image":[14],"captioning,":[15],"visual":[16],"question":[17],"answering":[18],"(VQA),":[19],"reasoning,":[20],"and":[21,60,72],"instruction":[22],"following.":[23],"Supporting":[24],"local":[25],"initiatives,":[27],"HordeVision":[28],"addresses":[29],"the":[30],"critical":[31],"need":[32],"Kazakh-language":[34],"multimodal":[35],"AI":[36],"systems.":[37],"Our":[38],"dataset":[39],"was":[40],"collected":[41],"using":[42,78],"a":[43,64,86],"syntactic":[44],"data":[45],"generation":[46],"pipeline":[47],"covering":[48],"diverse":[49],"contexts":[50],"including":[51,91],"social":[52],"life,":[53],"education,":[54],"work,":[55],"economy,":[56],"media,":[57],"communications,":[58],"culture,":[59],"heritage.":[61],"employ":[63],"two-stage":[65],"training":[66],"approach":[67],"combining":[68],"supervised":[69],"fine-tuning":[70],"(SFT)":[71],"reinforcement":[73],"learning":[74],"from":[75],"human":[76],"feedback":[77],"Group":[79],"Relative":[80],"Policy":[81],"Optimization":[82],"(GRPO).":[83],"Evaluated":[84],"on":[85],"human-annotated":[87],"benchmark":[88],"against":[89],"models":[90],"Google":[92],"Gemma,":[93]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-01-10T00:00:00"}
