{"id":"https://openalex.org/W7167604263","doi":"https://doi.org/10.48550/arxiv.2607.03328","title":"Beyond Post-Quantization: Native Hash Learning with a Dedicated HASH Token","display_name":"Beyond Post-Quantization: Native Hash Learning with a Dedicated HASH Token","publication_year":2026,"publication_date":"2026-07-03","ids":{"openalex":"https://openalex.org/W7167604263","doi":"https://doi.org/10.48550/arxiv.2607.03328"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2607.03328","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2607.03328","pdf_url":"https://arxiv.org/pdf/2607.03328","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2607.03328","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5023917479","display_name":"Xinze Liu","orcid":"https://orcid.org/0009-0006-4482-8995","observed_orcids":["https://orcid.org/0009-0006-4482-8995","https://orcid.org/0000-0002-3786-2646"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Xinze","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133331301","display_name":"Ding Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Ding","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002436544","display_name":"Dayan Wu","orcid":"https://orcid.org/0000-0002-8604-7226","observed_orcids":["https://orcid.org/0000-0002-8604-7226"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Dayan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135142094","display_name":"Hengjie Zhu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Hengjie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119959312","display_name":"Peng Fu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fu, Peng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148437336","display_name":"Zheng Lin","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Zheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5143443705","display_name":"Weiping Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Weiping","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":true,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9976000189781189,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9976000189781189,"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/T10824","display_name":"Image Retrieval and Classification Techniques","score":0.0012000000569969416,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0010999999940395355,"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/deep-hashing","display_name":"deep hashing","score":0.8420000076293945},{"id":"https://openalex.org/keywords/binary-code-generation","display_name":"binary code generation","score":0.46299999952316284},{"id":"https://openalex.org/keywords/vision-transformer","display_name":"vision transformer","score":0.753000020980835},{"id":"https://openalex.org/keywords/image-retrieval","display_name":"image retrieval","score":0.6100000143051147}],"concepts":[{"id":"https://openalex.org/C99138194","wikidata":"https://www.wikidata.org/wiki/Q183427","display_name":"Hash function","level":2,"score":0.7742999792098999},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7134000062942505},{"id":"https://openalex.org/C138111711","wikidata":"https://www.wikidata.org/wiki/Q478351","display_name":"Double hashing","level":4,"score":0.6669999957084656},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.5397999882698059},{"id":"https://openalex.org/C36375716","wikidata":"https://www.wikidata.org/wiki/Q6553456","display_name":"Linear hashing","level":5,"score":0.5394999980926514},{"id":"https://openalex.org/C122907437","wikidata":"https://www.wikidata.org/wiki/Q5318999","display_name":"Dynamic perfect hashing","level":5,"score":0.536899983882904},{"id":"https://openalex.org/C116058348","wikidata":"https://www.wikidata.org/wiki/Q846912","display_name":"Universal hashing","level":5,"score":0.5070000290870667},{"id":"https://openalex.org/C67388219","wikidata":"https://www.wikidata.org/wiki/Q207440","display_name":"Hash table","level":3,"score":0.4997999966144562},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.4959000051021576},{"id":"https://openalex.org/C2779494224","wikidata":"https://www.wikidata.org/wiki/Q5645799","display_name":"Hamming space","level":5,"score":0.4555000066757202},{"id":"https://openalex.org/C133667856","wikidata":"https://www.wikidata.org/wiki/Q5439682","display_name":"Feature hashing","level":5,"score":0.4165000021457672},{"id":"https://openalex.org/C63435697","wikidata":"https://www.wikidata.org/wiki/Q864135","display_name":"Binary code","level":3,"score":0.38499999046325684},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.33230000734329224},{"id":"https://openalex.org/C135783594","wikidata":"https://www.wikidata.org/wiki/Q5678864","display_name":"Hash chain","level":3,"score":0.3285999894142151},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.2782000005245209},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.25949999690055847}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2607.03328","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2607.03328","pdf_url":"https://arxiv.org/pdf/2607.03328","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2607.03328","doi":"https://doi.org/10.48550/arxiv.2607.03328","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.03328","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2607.03328","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2607.03328","pdf_url":"https://arxiv.org/pdf/2607.03328","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7167604263.pdf","grobid_xml":"https://content.openalex.org/works/W7167604263.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7213446822","https://openalex.org/W4296344225","https://openalex.org/W4225654135","https://openalex.org/W3159460504","https://openalex.org/W4392107350","https://openalex.org/W4406553827","https://openalex.org/W4417131199","https://openalex.org/W4385828981","https://openalex.org/W4407852416","https://openalex.org/W2920852583"],"abstract_inverted_index":{"Efficient":[0],"large-scale":[1],"image":[2],"retrieval":[3,110,229],"requires":[4],"compact":[5,211,236],"representations":[6,165],"that":[7,104,152,190,222],"preserve":[8],"semantic":[9,193],"similarity":[10,197],"under":[11],"fast":[12],"Hamming-space":[13],"search.":[14],"Deep":[15],"hashing":[16,93],"is":[17,118,184,240],"appealing,":[18],"but":[19],"most":[20],"existing":[21],"CNN-":[22],"and":[23,38,66,130,199,210],"ViT-based":[24],"methods":[25],"still":[26],"follow":[27],"a":[28,45,57,81,95,100,107,122,131,147,162,187],"post-quantization":[29],"paradigm,":[30],"where":[31],"continuous":[32,137],"visual":[33],"features":[34],"are":[35,41,166],"first":[36],"learned":[37],"binary":[39,127,212],"codes":[40],"then":[42],"produced":[43],"by":[44,179],"terminal":[46,96],"hash":[47,87],"projection":[48],"or":[49,226],"binarization":[50],"operation.":[51],"This":[52],"late":[53],"code":[54,128],"generation":[55,129],"creates":[56],"feature-to-code":[58],"discrepancy":[59],"between":[60],"the":[61,67,113,155,172,203,233],"continuously":[62],"optimized":[63,185],"representation":[64],"space":[65,70],"discrete":[68],"Hamming":[69,237],"used":[71,219],"for":[72,85,125,134],"retrieval.":[73],"To":[74,139],"address":[75],"this":[76],"limitation,":[77],"we":[78,144],"propose":[79],"HashViT,":[80],"Vision":[82],"Transformer":[83],"framework":[84],"native":[86],"token":[88,103,117,169,205],"learning.":[89],"Instead":[90],"of":[91,235],"treating":[92],"as":[94,106],"readout,":[97],"HashViT":[98,183,223],"introduces":[99],"dedicated":[101],"HASH":[102,116,204],"serves":[105],"persistent,":[108],"hash-oriented":[109],"state":[111],"inside":[112],"transformer.":[114],"The":[115],"structurally":[119],"decomposed":[120],"into":[121],"Hash":[123,149,156],"Register":[124,157],"direct":[126],"Semantic":[132],"Workspace":[133],"preserving":[135,232],"auxiliary":[136],"semantics.":[138],"enable":[140],"effective":[141],"workspace-to-register":[142],"interaction,":[143],"further":[145],"design":[146],"lightweight":[148],"Refinement":[150],"Adapter":[151],"progressively":[153],"refines":[154],"across":[158],"transformer":[159],"layers.":[160],"As":[161],"result,":[163],"binary-oriented":[164],"formed":[167],"through":[168],"evolution":[170],"within":[171],"backbone,":[173],"rather":[174],"than":[175],"being":[176],"abruptly":[177],"induced":[178],"an":[180],"output-level":[181],"projection.":[182],"with":[186],"unified":[188],"objective":[189],"combines":[191],"learnable":[192],"center":[194],"supervision,":[195],"class-token":[196],"distillation,":[198],"quantization":[200],"regularization,":[201],"encouraging":[202],"to":[206],"encode":[207],"semantically":[208],"structured":[209],"representations.":[213],"Extensive":[214],"experiments":[215],"on":[216],"three":[217],"widely":[218],"benchmarks":[220],"demonstrate":[221],"achieves":[224],"state-of-the-art":[225],"highly":[227],"competitive":[228],"performance":[230],"while":[231],"efficiency":[234],"codes.":[238],"Code":[239],"available":[241],"at":[242],"https://github.com/Xinze919/HashViT.":[243]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-07-08T00:00:00"}