{"id":"https://openalex.org/W7135221998","doi":"https://doi.org/10.48550/arxiv.2603.12265","title":"OmniStream: Mastering Perception, Reconstruction and Action in Continuous Streams","display_name":"OmniStream: Mastering Perception, Reconstruction and Action in Continuous Streams","publication_year":2026,"publication_date":"2026-03-12","ids":{"openalex":"https://openalex.org/W7135221998","doi":"https://doi.org/10.48550/arxiv.2603.12265"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.12265","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.12265","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2603.12265","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129039007","display_name":"Yibin Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan, Yibin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129020588","display_name":"Jilan Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Jilan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007532959","display_name":"Shangzhe Di","orcid":"https://orcid.org/0009-0005-6977-4332"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Di, Shangzhe","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039135087","display_name":"Haoning Wu","orcid":"https://orcid.org/0009-0001-8717-338X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Haoning","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5053549250","display_name":"Weidi Xie","orcid":"https://orcid.org/0009-0002-8609-6826"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Weidi","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":0.8217999935150146,"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":0.8217999935150146,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.0908999964594841,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.01489999983459711,"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/representation","display_name":"Representation (politics)","score":0.5138000249862671},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.5083000063896179},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.42410001158714294},{"id":"https://openalex.org/keywords/embodied-cognition","display_name":"Embodied cognition","score":0.4196999967098236},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.3659999966621399},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.3614000082015991},{"id":"https://openalex.org/keywords/robotics","display_name":"Robotics","score":0.3490000069141388},{"id":"https://openalex.org/keywords/point","display_name":"Point (geometry)","score":0.3479999899864197}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7964000105857849},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5789999961853027},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5138000249862671},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.5083000063896179},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4805000126361847},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.4375},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.42410001158714294},{"id":"https://openalex.org/C100609095","wikidata":"https://www.wikidata.org/wiki/Q1335050","display_name":"Embodied cognition","level":2,"score":0.4196999967098236},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3659999966621399},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3614000082015991},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.3490000069141388},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.3479999899864197},{"id":"https://openalex.org/C193611912","wikidata":"https://www.wikidata.org/wiki/Q4677596","display_name":"Active vision","level":2,"score":0.3149999976158142},{"id":"https://openalex.org/C160086991","wikidata":"https://www.wikidata.org/wiki/Q5939193","display_name":"Human visual system model","level":3,"score":0.3003999888896942},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.3000999987125397},{"id":"https://openalex.org/C16345878","wikidata":"https://www.wikidata.org/wiki/Q107472979","display_name":"Orientation (vector space)","level":2,"score":0.295199990272522},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.28279998898506165},{"id":"https://openalex.org/C105842133","wikidata":"https://www.wikidata.org/wiki/Q1899679","display_name":"Visual communication","level":2,"score":0.2808000147342682},{"id":"https://openalex.org/C89198739","wikidata":"https://www.wikidata.org/wiki/Q3079880","display_name":"Data stream mining","level":2,"score":0.2732999920845032},{"id":"https://openalex.org/C2778251979","wikidata":"https://www.wikidata.org/wiki/Q7936617","display_name":"Visual processing","level":3,"score":0.27129998803138733},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.26919999718666077},{"id":"https://openalex.org/C5339829","wikidata":"https://www.wikidata.org/wiki/Q1425977","display_name":"Machine vision","level":2,"score":0.25270000100135803}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.12265","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.12265","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2603.12265","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.12265","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Modern":[0],"visual":[1,44,54,175],"agents":[2],"require":[3],"representations":[4],"that":[5,46,160],"are":[6],"general,":[7],"causal,":[8],"and":[9,50,61,91,98,124,132,165,179],"physically":[10],"structured":[11],"to":[12],"operate":[13],"in":[14,27],"real-time":[15],"streaming":[16,43,95,127],"environments.":[17],"However,":[18],"current":[19],"vision":[20,158],"foundation":[21],"models":[22],"remain":[23],"fragmented,":[24],"specializing":[25],"narrowly":[26],"image":[28,123],"semantic":[29],"perception,":[30],"offline":[31],"temporal":[32,92,166],"modeling,":[33],"or":[34],"spatial":[35,133],"geometry.":[36],"This":[37],"paper":[38],"introduces":[39],"OmniStream,":[40],"a":[41,78,85,110,155,169],"unified":[42],"backbone":[45,159],"effectively":[47],"perceives,":[48],"reconstructs,":[49],"acts":[51],"from":[52],"diverse":[53],"inputs.":[55],"By":[56],"incorporating":[57],"causal":[58],"spatiotemporal":[59],"attention":[60],"3D":[62],"rotary":[63],"positional":[64],"embeddings":[65],"(3D-RoPE),":[66],"our":[67,148],"model":[68],"supports":[69],"efficient,":[70],"frame-by-frame":[71],"online":[72],"processing":[73],"of":[74,153],"video":[75,125,131],"streams":[76],"via":[77],"persistent":[79],"KV-cache.":[80],"We":[81],"pre-train":[82],"OmniStream":[83,114],"using":[84],"synergistic":[86],"multi-task":[87],"framework":[88],"coupling":[89],"static":[90],"representation":[93],"learning,":[94],"geometric":[96,128],"reconstruction,":[97,129],"vision-language":[99],"alignment":[100],"on":[101],"29":[102],"datasets.":[103],"Extensive":[104],"evaluations":[105],"show":[106],"that,":[107],"even":[108],"with":[109,119],"strictly":[111],"frozen":[112],"backbone,":[113],"achieves":[115],"consistently":[116],"competitive":[117],"performance":[118],"specialized":[120],"experts":[121],"across":[122,162],"probing,":[126],"complex":[130],"reasoning,":[134,167],"as":[135,137],"well":[136],"robotic":[138],"manipulation":[139],"(unseen":[140],"at":[141],"training).":[142],"Rather":[143],"than":[144],"pursuing":[145],"benchmark-specific":[146],"dominance,":[147],"work":[149],"demonstrates":[150],"the":[151],"viability":[152],"training":[154],"single,":[156],"versatile":[157],"generalizes":[161],"semantic,":[163],"spatial,":[164],"i.e.,":[168],"more":[170],"meaningful":[171],"step":[172],"toward":[173],"general-purpose":[174],"understanding":[176],"for":[177],"interactive":[178],"embodied":[180],"agents.":[181]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-14T00:00:00"}