{"id":"https://openalex.org/W7214728559","doi":"https://doi.org/10.48550/arxiv.2609.32699","title":"RandSlot: Learning Compact Visual Document Representations with Random Soft Tokens","display_name":"RandSlot: Learning Compact Visual Document Representations with Random Soft Tokens","publication_year":2026,"publication_date":"2026-09-26","ids":{"openalex":"https://openalex.org/W7214728559","doi":"https://doi.org/10.48550/arxiv.2609.32699"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.32699","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.32699","pdf_url":"https://arxiv.org/pdf/2609.32699","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","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.32699","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5153403574","display_name":"Dewen Guo","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405270328","display_name":"Ant Group (China)","ror":"https://ror.org/00g5w7n47","country_code":"CN","type":"company","lineage":["https://openalex.org/I4405270328"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guo, Dewen","raw_affiliation_strings":["Ant Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ant Group","institution_ids":["https://openalex.org/I4405270328"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153444386","display_name":"Shi Yu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu, Shi","raw_affiliation_strings":["Tsinghua University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153486703","display_name":"Lingxiao Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405270328","display_name":"Ant Group (China)","ror":"https://ror.org/00g5w7n47","country_code":"CN","type":"company","lineage":["https://openalex.org/I4405270328"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhang, Lingxiao","raw_affiliation_strings":["Ant Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ant Group","institution_ids":["https://openalex.org/I4405270328"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153454099","display_name":"Yang Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405270328","display_name":"Ant Group (China)","ror":"https://ror.org/00g5w7n47","country_code":"CN","type":"company","lineage":["https://openalex.org/I4405270328"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhang, Yang","raw_affiliation_strings":["Ant Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ant Group","institution_ids":["https://openalex.org/I4405270328"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153456912","display_name":"Tao XU","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405270328","display_name":"Ant Group (China)","ror":"https://ror.org/00g5w7n47","country_code":"CN","type":"company","lineage":["https://openalex.org/I4405270328"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"XU, Tao","raw_affiliation_strings":["Ant Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ant Group","institution_ids":["https://openalex.org/I4405270328"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5153460849","display_name":"Dan Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405270328","display_name":"Ant Group (China)","ror":"https://ror.org/00g5w7n47","country_code":"CN","type":"company","lineage":["https://openalex.org/I4405270328"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Dan","raw_affiliation_strings":["Ant Group"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ant Group","institution_ids":["https://openalex.org/I4405270328"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.6689000129699707,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.6689000129699707,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.12189999967813492,"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/T10028","display_name":"Topic Modeling","score":0.08060000091791153,"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/zero-shot-inference","display_name":"zero-shot inference","score":0.5569999814033508},{"id":"https://openalex.org/keywords/tables","display_name":"tables","score":0.5559999942779541}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7459999918937683},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6491000056266785},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5788000226020813},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.4848000109195709},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4846999943256378},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4693000018596649},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.44020000100135803},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.4332999885082245},{"id":"https://openalex.org/C161156560","wikidata":"https://www.wikidata.org/wiki/Q1638872","display_name":"Document retrieval","level":2,"score":0.3801000118255615},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.36970001459121704},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.35589998960494995},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.3431999981403351},{"id":"https://openalex.org/C1667742","wikidata":"https://www.wikidata.org/wiki/Q10927554","display_name":"Image retrieval","level":3,"score":0.3179999887943268},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.3084000051021576},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.29190000891685486},{"id":"https://openalex.org/C75165309","wikidata":"https://www.wikidata.org/wiki/Q2258979","display_name":"Search engine indexing","level":2,"score":0.2904999852180481},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2822999954223633},{"id":"https://openalex.org/C159254197","wikidata":"https://www.wikidata.org/wiki/Q1144915","display_name":"Lexicographical order","level":2,"score":0.28189998865127563},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.27970001101493835},{"id":"https://openalex.org/C189391414","wikidata":"https://www.wikidata.org/wiki/Q7936579","display_name":"Visual Word","level":4,"score":0.2639000117778778},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.25760000944137573},{"id":"https://openalex.org/C89686163","wikidata":"https://www.wikidata.org/wiki/Q1187982","display_name":"Vector space model","level":2,"score":0.257099986076355},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2549999952316284}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.32699","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.32699","pdf_url":"https://arxiv.org/pdf/2609.32699","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.2609.32699","doi":"https://doi.org/10.48550/arxiv.2609.32699","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.32699","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:2609.32699","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.32699","pdf_url":"https://arxiv.org/pdf/2609.32699","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":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7214728559.pdf","grobid_xml":"https://content.openalex.org/works/W7214728559.grobid-xml"},"referenced_works_count":27,"referenced_works":["https://openalex.org/W2111406701","https://openalex.org/W3034439313","https://openalex.org/W3152956381","https://openalex.org/W3156636935","https://openalex.org/W3168867926","https://openalex.org/W3174770825","https://openalex.org/W4226518707","https://openalex.org/W4297808394","https://openalex.org/W4399251701","https://openalex.org/W4404356281","https://openalex.org/W4405254469","https://openalex.org/W4407794264","https://openalex.org/W4413146941","https://openalex.org/W4414689452","https://openalex.org/W4415163284","https://openalex.org/W4415251124","https://openalex.org/W4415334930","https://openalex.org/W4416780398","https://openalex.org/W7083825781","https://openalex.org/W7120003758","https://openalex.org/W7123399714","https://openalex.org/W7126206008","https://openalex.org/W7133201354","https://openalex.org/W7154609584","https://openalex.org/W7161000475","https://openalex.org/W7166804103","https://openalex.org/W7167899984"],"related_works":["https://openalex.org/W7160626780","https://openalex.org/W7167918601","https://openalex.org/W7163077403","https://openalex.org/W7172397956","https://openalex.org/W7223278214","https://openalex.org/W7212306526","https://openalex.org/W7129105573","https://openalex.org/W7202209665","https://openalex.org/W7154397759","https://openalex.org/W7214731532"],"abstract_inverted_index":{"Visual":[0],"document":[1,46,63],"retrieval":[2,31,93,119,157],"requires":[3,164],"expressive":[4],"representations":[5,19,47,158],"to":[6,42,60,87,102],"match":[7],"queries":[8],"with":[9,48,83,111,144],"evidence":[10],"distributed":[11],"across":[12],"text,":[13],"tables,":[14],"and":[15,25,62,66],"page":[16],"layouts.":[17],"Multi-vector":[18],"capture":[20],"fine-grained":[21],"information,":[22],"but":[23],"storing":[24],"comparing":[26],"many":[27],"vectors":[28,59],"introduces":[29],"substantial":[30],"costs.":[32],"In":[33],"this":[34],"paper,":[35],"we":[36,54],"introduce":[37],"RandSlot,":[38],"a":[39,89],"simple":[40],"approach":[41],"learn":[43],"compact":[44,156],"visual":[45],"random":[49,57,139,150],"soft":[50,140],"tokens.":[51],"During":[52],"training,":[53],"append":[55],"independently-sampled":[56],"unit":[58],"query":[61],"input":[64,108],"sequences":[65],"resample":[67],"them":[68],"at":[69,146],"every":[70],"use,":[71],"without":[72],"introducing":[73],"learnable":[74],"soft-token":[75],"parameters.":[76],"The":[77],"encoder":[78,101],"contextualizes":[79],"these":[80,134],"auxiliary":[81],"inputs":[82,151],"the":[84,100,126],"original":[85],"content":[86],"produce":[88],"small":[90],"set":[91],"of":[92],"vectors.":[94],"A":[95],"standard":[96],"late-interaction":[97],"objective":[98],"trains":[99],"extract":[103],"relevant":[104],"information":[105],"under":[106,125],"varying":[107],"conditions.":[109],"Experiments":[110],"different":[112],"backbone":[113],"models":[114],"show":[115],"that":[116,133,149],"RandSlot":[117],"improves":[118],"quality":[120],"over":[121],"alternative":[122],"readout":[123],"strategies":[124],"same":[127],"vector":[128],"budget.":[129],"Further":[130],"analysis":[131],"shows":[132],"gains":[135],"can":[136,154],"persist":[137],"when":[138,160],"tokens":[141],"are":[142],"replaced":[143],"zeros":[145],"inference,":[147],"demonstrating":[148],"during":[152],"training":[153],"improve":[155],"even":[159],"inference":[161],"no":[162],"longer":[163],"sampling.":[165]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-30T00:00:00"}