{"id":"https://openalex.org/W7167061089","doi":"https://doi.org/10.48550/arxiv.2607.00465","title":"StochasT: Learning with Stochastic Turn Depth for Visual Instruction Tuning","display_name":"StochasT: Learning with Stochastic Turn Depth for Visual Instruction Tuning","publication_year":2026,"publication_date":"2026-07-01","ids":{"openalex":"https://openalex.org/W7167061089","doi":"https://doi.org/10.48550/arxiv.2607.00465"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2607.00465","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.00465","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2607.00465","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139917076","display_name":"Yuan Qing","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qing, Yuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061451821","display_name":"Chengzhi Mao","orcid":"https://orcid.org/0009-0008-7791-3891"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mao, Chengzhi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5039597152","display_name":"B Dianxuan Gong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gong, Boqing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"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":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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9649999737739563,"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":0.9649999737739563,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.006599999964237213,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.005200000014156103,"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/overfitting","display_name":"Overfitting","score":0.7476000189781189},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6252999901771545},{"id":"https://openalex.org/keywords/dropout","display_name":"Dropout (neural networks)","score":0.5958999991416931},{"id":"https://openalex.org/keywords/measure","display_name":"Measure (data warehouse)","score":0.3917999863624573},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.34119999408721924},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.32519999146461487}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7692000269889832},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.7476000189781189},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.631600022315979},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6252999901771545},{"id":"https://openalex.org/C2776145597","wikidata":"https://www.wikidata.org/wiki/Q25339462","display_name":"Dropout (neural networks)","level":2,"score":0.5958999991416931},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.46709999442100525},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.3917999863624573},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.34119999408721924},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.32519999146461487},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.3102000057697296},{"id":"https://openalex.org/C157524613","wikidata":"https://www.wikidata.org/wiki/Q2828883","display_name":"Fine-tuning","level":2,"score":0.28870001435279846},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.2865999937057495},{"id":"https://openalex.org/C3020716817","wikidata":"https://www.wikidata.org/wiki/Q4132092","display_name":"Visual feedback","level":2,"score":0.2583000063896179},{"id":"https://openalex.org/C8272713","wikidata":"https://www.wikidata.org/wiki/Q176737","display_name":"Stochastic process","level":2,"score":0.2540999948978424}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2607.00465","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.00465","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2607.00465","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.00465","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.7415550947189331}],"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],"Vision-Language":[1],"Models":[2],"(LVLMs)":[3],"rely":[4],"extensively":[5],"on":[6,111,141],"Visual":[7],"Instruction":[8],"Tuning":[9],"(VIT)":[10],"to":[11,64,124,146],"elicit":[12],"their":[13,66,104],"multimodal":[14],"reasoning":[15],"capabilities.":[16],"However,":[17],"we":[18,78,133],"find":[19],"a":[20,135],"discrepancy:":[21],"VIT":[22],"often":[23],"packs":[24],"multiple":[25],"language":[26,89],"tasks":[27,90],"about":[28],"the":[29,70,76,92,126,129,142],"same":[30,93],"image":[31,94],"for":[32,62,91,116,165],"conversational,":[33],"multi-turn":[34,57,169],"training,":[35,58],"whereas":[36],"existing":[37],"benchmarks":[38],"evaluate":[39],"LVLMs":[40,161],"in":[41,69],"isolated,":[42],"single-turn":[43,167],"scenarios.":[44],"The":[45],"models":[46],"can":[47],"suffer":[48],"from":[49],"visual":[50],"attention":[51],"decay":[52],"and":[53,113,168],"contextual":[54,152],"overfitting":[55],"during":[56],"making":[59],"it":[60,118],"hard":[61],"them":[63],"realize":[65],"full":[67],"potential":[68],"mismatched":[71],"test":[72],"phase.":[73],"To":[74],"close":[75],"gap,":[77],"propose":[79],"learning":[80],"with":[81],"Stochastic":[82],"Turn":[83],"Depth":[84],"(StochasT),":[85],"which":[86],"stochastically":[87],"groups":[88],"into":[95],"clusters":[96],"of":[97,128],"varying":[98,151],"sizes":[99],"(turn":[100],"depth)":[101],"while":[102,108],"preserving":[103],"organic":[105],"order.":[106],"Hence,":[107],"StochasT":[109,158],"draws":[110],"Dropout":[112],"stochastic":[114],"depth":[115],"ResNets,":[117],"does":[119],"not":[120],"actually":[121],"drop":[122],"anything":[123],"maximize":[125],"utility":[127],"training":[130],"data.":[131],"Furthermore,":[132],"introduce":[134],"challenging,":[136],"benchmark-agnostic":[137],"evaluation":[138],"mechanism":[139],"based":[140],"Balanced":[143],"Latin":[144],"Square":[145],"measure":[147],"LVLMs'":[148],"robustness":[149],"under":[150],"dependencies.":[153],"Extensive":[154],"experiments":[155],"demonstrate":[156],"that":[157],"effectively":[159],"grants":[160],"strong,":[162],"harmonized":[163],"capabilities":[164],"both":[166],"use":[170],"cases.":[171]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-07-03T00:00:00"}