{"id":"https://openalex.org/W7116841178","doi":"https://doi.org/10.48550/arxiv.2512.17221","title":"DAVE: A VLM Vision Encoder for Document Understanding and Web Agents","display_name":"DAVE: A VLM Vision Encoder for Document Understanding and Web Agents","publication_year":2025,"publication_date":"2025-12-19","ids":{"openalex":"https://openalex.org/W7116841178","doi":"https://doi.org/10.48550/arxiv.2512.17221"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2512.17221","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.17221","pdf_url":"https://arxiv.org/pdf/2512.17221","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/2512.17221","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5111084088","display_name":"Brandon Huang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Brandon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102027020","display_name":"Hang Hua","orcid":"https://orcid.org/0000-0002-5441-5776","observed_orcids":["https://orcid.org/0000-0002-5441-5776"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hua, Hang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121071445","display_name":"Zhuoran Yu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Zhuoran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029105520","display_name":"Trevor Darrell","orcid":"https://orcid.org/0000-0001-5453-8533","observed_orcids":["https://orcid.org/0000-0001-5453-8533"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Darrell, Trevor","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111930150","display_name":"Rogerio S. Feris","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Feris, Rogerio","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5066997569","display_name":"Roei Herzig","orcid":"https://orcid.org/0000-0002-5451-6059","observed_orcids":["https://orcid.org/0000-0002-5451-6059"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Herzig, Roei","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":0.9965999722480774,"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":0.9965999722480774,"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/T10601","display_name":"Handwritten Text Recognition Techniques","score":0.003000000026077032,"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":9.999999747378752e-05,"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/vision-encoder","display_name":"vision encoder","score":0.8119999766349792},{"id":"https://openalex.org/keywords/vision-language-models","display_name":"vision-language models","score":0.8989999890327454},{"id":"https://openalex.org/keywords/document-understanding","display_name":"document understanding","score":0.9959999918937683},{"id":"https://openalex.org/keywords/web-agents","display_name":"web agents","score":0.9670000076293945},{"id":"https://openalex.org/keywords/self-supervised-pretraining","display_name":"self-supervised pre-training","score":0.8849999904632568},{"id":"https://openalex.org/keywords/unlabeled-data","display_name":"unlabeled data","score":0.640999972820282},{"id":"https://openalex.org/keywords/model-merging","display_name":"model merging","score":0.703000009059906},{"id":"https://openalex.org/keywords/ensemble-training","display_name":"ensemble training","score":0.7919999957084656},{"id":"https://openalex.org/keywords/document-parsing","display_name":"document parsing","score":0.46700000762939453},{"id":"https://openalex.org/keywords/visual-question-answering","display_name":"visual question answering","score":0.8190000057220459}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7924000024795532},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.7020000219345093},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6345999836921692},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5745999813079834},{"id":"https://openalex.org/C186644900","wikidata":"https://www.wikidata.org/wiki/Q194152","display_name":"Parsing","level":2,"score":0.5619999766349792},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5196999907493591},{"id":"https://openalex.org/C118643609","wikidata":"https://www.wikidata.org/wiki/Q189210","display_name":"Web application","level":2,"score":0.42899999022483826},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.42170000076293945},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.4124999940395355},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.38089999556541443},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.35850000381469727},{"id":"https://openalex.org/C62230096","wikidata":"https://www.wikidata.org/wiki/Q275969","display_name":"Crowdsourcing","level":2,"score":0.33410000801086426},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3037000000476837},{"id":"https://openalex.org/C21959979","wikidata":"https://www.wikidata.org/wiki/Q36774","display_name":"Web page","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.28360000252723694},{"id":"https://openalex.org/C35578498","wikidata":"https://www.wikidata.org/wiki/Q193424","display_name":"Web service","level":2,"score":0.28279998898506165},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.2732999920845032},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.26589998602867126},{"id":"https://openalex.org/C141353440","wikidata":"https://www.wikidata.org/wiki/Q182221","display_name":"Fuse (electrical)","level":2,"score":0.25699999928474426}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2512.17221","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.17221","pdf_url":"https://arxiv.org/pdf/2512.17221","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.2512.17221","doi":"https://doi.org/10.48550/arxiv.2512.17221","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.17221","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2512.17221","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.17221","pdf_url":"https://arxiv.org/pdf/2512.17221","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":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4415797672","https://openalex.org/W4407359580","https://openalex.org/W4403808856","https://openalex.org/W4402716457","https://openalex.org/W4403842477","https://openalex.org/W7163523875","https://openalex.org/W7101645830","https://openalex.org/W6948243661","https://openalex.org/W4413259061","https://openalex.org/W7103104106"],"abstract_inverted_index":{"While":[0],"Vision-language":[1],"models":[2],"(VLMs)":[3],"have":[4],"demonstrated":[5],"remarkable":[6],"performance":[7],"across":[8],"multi-modal":[9],"tasks,":[10,182],"their":[11,20],"choice":[12],"of":[13,192],"vision":[14,45,200],"encoders":[15,141,167],"presents":[16],"a":[17,44,81,90,136,198],"fundamental":[18],"weakness:":[19],"low-level":[21],"features":[22,163],"lack":[23],"the":[24,67,96,109,190],"robust":[25],"structural":[26],"and":[27,34,50,75,102,126,129,174,186,204],"spatial":[28],"information":[29],"essential":[30],"for":[31,48,52,69,73,202],"document":[32,74,128,173,181,203],"understanding":[33],"web":[35,76,130,153,184,205],"agents.":[36],"To":[37],"bridge":[38],"this":[39],"gap,":[40],"we":[41,112],"introduce":[42,135],"DAVE,":[43],"encoder":[46,201],"purpose-built":[47],"VLMs":[49],"tailored":[51],"these":[53],"tasks.":[54],"Our":[55],"training":[56,160],"pipeline":[57],"is":[58],"designed":[59],"to":[60,65,116,147,161],"leverage":[61],"abundant":[62],"unlabeled":[63,86],"data":[64],"bypass":[66],"need":[68],"costly":[70],"large-scale":[71],"annotations":[72],"images.":[77],"We":[78,134,157],"begin":[79],"with":[80,121,143,151,170],"self-supervised":[82],"pretraining":[83,93],"stage":[84],"on":[85,179],"images,":[87],"followed":[88],"by":[89],"supervised":[91,110],"autoregressive":[92],"stage,":[94,111],"where":[95],"model":[97],"learns":[98],"tasks":[99],"like":[100],"parsing":[101],"localization":[103],"from":[104,164],"limited,":[105],"high-quality":[106],"data.":[107],"Within":[108],"adopt":[113],"two":[114],"strategies":[115],"improve":[117],"our":[118,171,193],"encoder's":[119],"alignment":[120],"both":[122],"general":[123],"visual":[124],"knowledge":[125],"diverse":[127],"agentic":[131,154],"tasks:":[132],"(i)":[133],"novel":[137],"model-merging":[138],"scheme,":[139],"combining":[140],"trained":[142],"different":[144,152],"text":[145],"decoders":[146],"ensure":[148],"broad":[149],"compatibility":[150],"architectures.":[155],"(ii)":[156],"use":[158],"ensemble":[159],"fuse":[162],"pretrained":[165],"generalist":[166],"(e.g.,":[168],"SigLIP2)":[169],"own":[172],"web-specific":[175],"representations.":[176],"Extensive":[177],"experiments":[178],"classic":[180],"VQAs,":[183],"localization,":[185],"agent-based":[187],"benchmarks":[188],"validate":[189],"effectiveness":[191],"approach,":[194],"establishing":[195],"DAVE":[196],"as":[197],"strong":[199],"applications.":[206]},"counts_by_year":[],"updated_date":"2026-10-30T12:39:25.321706","created_date":"2025-12-23T00:00:00"}