{"id":"https://openalex.org/W7165123881","doi":"https://doi.org/10.48550/arxiv.2606.18426","title":"VEGA: Learning Navigation VLAs from In-the-Wild Egocentric Video with Geometric Trajectory Supervision","display_name":"VEGA: Learning Navigation VLAs from In-the-Wild Egocentric Video with Geometric Trajectory Supervision","publication_year":2026,"publication_date":"2026-06-16","ids":{"openalex":"https://openalex.org/W7165123881","doi":"https://doi.org/10.48550/arxiv.2606.18426"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.18426","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.18426","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":"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.18426","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138863865","display_name":"Gershom Seneviratne","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Seneviratne, Gershom","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120724004","display_name":"Yohan Abeysinghe","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Abeysinghe, Yohan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060252105","display_name":"Jianyu An","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"An, Jianyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127874523","display_name":"Vaibhav Shende","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shende, Vaibhav","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138915754","display_name":"Dinesh Manocha","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Manocha, Dinesh","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":0.3871000111103058,"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":0.3871000111103058,"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/T10586","display_name":"Robotic Path Planning Algorithms","score":0.3312999904155731,"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/T11099","display_name":"Autonomous Vehicle Technology and Safety","score":0.05820000171661377,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/obstacle","display_name":"Obstacle","score":0.6766999959945679},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.663100004196167},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.6065000295639038},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5651000142097473},{"id":"https://openalex.org/keywords/usable","display_name":"USable","score":0.4586000144481659},{"id":"https://openalex.org/keywords/motion","display_name":"Motion (physics)","score":0.4359000027179718},{"id":"https://openalex.org/keywords/monocular","display_name":"Monocular","score":0.4230000078678131},{"id":"https://openalex.org/keywords/obstacle-avoidance","display_name":"Obstacle avoidance","score":0.38580000400543213},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.3425000011920929}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7121000289916992},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.7049999833106995},{"id":"https://openalex.org/C2776650193","wikidata":"https://www.wikidata.org/wiki/Q264661","display_name":"Obstacle","level":2,"score":0.6766999959945679},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.667900025844574},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.663100004196167},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.6065000295639038},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5651000142097473},{"id":"https://openalex.org/C2780615836","wikidata":"https://www.wikidata.org/wiki/Q2471869","display_name":"USable","level":2,"score":0.4586000144481659},{"id":"https://openalex.org/C104114177","wikidata":"https://www.wikidata.org/wiki/Q79782","display_name":"Motion (physics)","level":2,"score":0.4359000027179718},{"id":"https://openalex.org/C65909025","wikidata":"https://www.wikidata.org/wiki/Q1945033","display_name":"Monocular","level":2,"score":0.4230000078678131},{"id":"https://openalex.org/C6683253","wikidata":"https://www.wikidata.org/wiki/Q7075535","display_name":"Obstacle avoidance","level":4,"score":0.38580000400543213},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.3425000011920929},{"id":"https://openalex.org/C81074085","wikidata":"https://www.wikidata.org/wiki/Q366872","display_name":"Motion planning","level":3,"score":0.335999995470047},{"id":"https://openalex.org/C555745239","wikidata":"https://www.wikidata.org/wiki/Q133220","display_name":"Sonar","level":2,"score":0.3280999958515167},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.3156000077724457},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.3125999867916107},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.30790001153945923},{"id":"https://openalex.org/C121704057","wikidata":"https://www.wikidata.org/wiki/Q352070","display_name":"Collision","level":2,"score":0.30660000443458557},{"id":"https://openalex.org/C158829959","wikidata":"https://www.wikidata.org/wiki/Q1640606","display_name":"Monocular vision","level":2,"score":0.30649998784065247},{"id":"https://openalex.org/C2780864053","wikidata":"https://www.wikidata.org/wiki/Q5147495","display_name":"Collision avoidance","level":3,"score":0.2994999885559082},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.2985999882221222},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.2921999990940094},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2808000147342682},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.259799987077713},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.25769999623298645},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.25690001249313354}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.18426","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.18426","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.18426","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.18426","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":"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"},"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":{"We":[0],"introduce":[1,127],"VEGA,":[2],"an":[3],"approach":[4],"for":[5,46,219],"training":[6,220],"navigation":[7,14,59,79,108,138,222],"VisionLanguage-Action":[8],"(VLA)":[9],"models":[10],"from":[11,75],"unlabeled":[12],"egocentric":[13,17],"videos.":[15],"Internet-scale":[16],"videos":[18,41],"provide":[19,53],"a":[20,105,122,129,214],"scalable":[21,215],"source":[22],"of":[23,154,234],"navigation-relevant":[24],"visual":[25],"observations,":[26],"capturing":[27],"cluttered":[28],"scenes,":[29],"close-range":[30],"obstacles,":[31],"and":[32,88,134,151,170,195,216,226],"natural":[33],"human":[34],"motion":[35],"through":[36],"real-world":[37,204],"spaces.":[38],"However,":[39],"these":[40],"are":[42],"not":[43,52],"directly":[44,120],"usable":[45],"policy":[47],"learning":[48],"because":[49],"they":[50],"do":[51],"obstacle-aware":[54,90,118,221],"trajectories":[55,91],"conditioned":[56],"on":[57,180],"explicit":[58],"goals":[60,80,139],"in":[61,203],"the":[62,93,177,232],"robot's":[63],"coordinate":[64],"frame.":[65],"VEGA":[66,116,160],"addresses":[67],"this":[68],"gap":[69],"by":[70,168,174,185,191,199],"reconstructing":[71],"local":[72],"scene":[73,142],"geometry":[74,112],"monocular":[76],"video,":[77],"sampling":[78],"(represented":[81],"as":[82],"text,":[83],"image,":[84],"or":[85],"spatial":[86],"waypoints)":[87],"generating":[89],"using":[92,111],"constructed":[94],"geometry.":[95],"The":[96,224],"resulting":[97],"trajectory":[98],"distribution":[99],"is":[100],"then":[101],"used":[102],"to":[103,145],"train":[104],"flow-matching":[106],"VLA":[107],"policy.":[109,124],"By":[110],"exclusively":[113],"during":[114],"training,":[115],"distills":[117],"planning":[119],"into":[121],"vision-based":[123],"Furthermore,":[125],"we":[126,207],"VEGA-Bench,":[128],"benchmark":[130,227],"containing":[131],"250k":[132],"scenes":[133],"approximately":[135],"5":[136],"million":[137],"paired":[140],"with":[141],"geometry,":[143],"designed":[144],"evaluate":[146],"goal":[147,163],"progress,":[148],"collision":[149],"avoidance,":[150],"obstacle":[152,172,197],"clearance":[153,173,198],"VLAs.":[155,223],"Our":[156],"evaluation":[157],"shows":[158],"that":[159,209],"achieves":[161],"competitive":[162],"progress":[164],"while":[165,182],"reducing":[166,189],"collisions":[167,190],"33.0%":[169],"improving":[171,183,196],"17.9%":[175],"over":[176],"strongest":[178],"baseline":[179],"VEGABench,":[181],"success":[184],"at":[186,192,200,231],"least":[187,193,201],"150.0%,":[188],"66.7%,":[194],"60.0%":[202],"trials.":[205],"Ultimately,":[206],"demonstrate":[208],"video-derived":[210],"geometric":[211],"supervision":[212],"provides":[213],"effective":[217],"signal":[218],"code":[225],"will":[228],"be":[229],"released":[230],"time":[233],"publication.":[235]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-19T00:00:00"}