{"id":"https://openalex.org/W7130308819","doi":"https://doi.org/10.48550/arxiv.2602.15724","title":"Learning to Retrieve Navigable Candidates for Efficient Vision-and-Language Navigation","display_name":"Learning to Retrieve Navigable Candidates for Efficient Vision-and-Language Navigation","publication_year":2026,"publication_date":"2026-02-17","ids":{"openalex":"https://openalex.org/W7130308819","doi":"https://doi.org/10.48550/arxiv.2602.15724"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2602.15724","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.15724","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2602.15724","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5126339810","display_name":"Shutian Gu","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Gu, Shutian","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126342539","display_name":"Chengkai Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Chengkai","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126291868","display_name":"Ruoyu Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Ruoyu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5126279606","display_name":"Lina Yao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yao, Lina","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5126339810"],"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":0.9837999939918518,"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":0.9837999939918518,"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":0.005799999926239252,"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"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.002099999925121665,"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/ambiguity","display_name":"Ambiguity","score":0.6380000114440918},{"id":"https://openalex.org/keywords/oracle","display_name":"Oracle","score":0.5788999795913696},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5673999786376953},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.49149999022483826},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.483599990606308},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.4578000009059906},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.4462999999523163},{"id":"https://openalex.org/keywords/prior-probability","display_name":"Prior probability","score":0.4453999996185303}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8026999831199646},{"id":"https://openalex.org/C2780522230","wikidata":"https://www.wikidata.org/wiki/Q1140419","display_name":"Ambiguity","level":2,"score":0.6380000114440918},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.583899974822998},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.5788999795913696},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5673999786376953},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.49149999022483826},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.483599990606308},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.4578000009059906},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.4462999999523163},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.4453999996185303},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.414900004863739},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.41100001335144043},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.3765999972820282},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.35839998722076416},{"id":"https://openalex.org/C2779182362","wikidata":"https://www.wikidata.org/wiki/Q17126187","display_name":"Session (web analytics)","level":2,"score":0.3447999954223633},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.31060001254081726},{"id":"https://openalex.org/C2779606619","wikidata":"https://www.wikidata.org/wiki/Q17092524","display_name":"Interchangeability","level":2,"score":0.31029999256134033},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3089999854564667},{"id":"https://openalex.org/C2775955345","wikidata":"https://www.wikidata.org/wiki/Q7449071","display_name":"Semantic mapping","level":2,"score":0.2879999876022339},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2791999876499176},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.2775999903678894},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.27549999952316284},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2750000059604645},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.25760000944137573}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2602.15724","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.15724","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2602.15724","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.15724","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":"article"},"sustainable_development_goals":[{"score":0.4561276435852051,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Vision-and-Language":[0],"Navigation":[1],"(VLN)":[2],"requires":[3],"an":[4,100,123,209],"agent":[5],"to":[6,28,71,194],"follow":[7],"natural-language":[8],"instructions":[9,49],"and":[10,31,52,56,75,137,146,172,177,188,197,211],"navigate":[11],"through":[12],"previously":[13],"unseen":[14,178],"environments.":[15,179],"Recent":[16],"approaches":[17],"increasingly":[18],"employ":[19],"large":[20],"language":[21,86],"models":[22],"(LLMs)":[23],"as":[24,43,110],"high-level":[25],"navigators":[26],"due":[27],"their":[29],"flexibility":[30],"reasoning":[32],"capability.":[33],"However,":[34],"prompt-based":[35],"LLM":[36,132],"navigation":[37,108],"often":[38],"suffers":[39],"from":[40,50],"inefficient":[41],"decision-making,":[42],"the":[44,73,84,97,120,150,157],"model":[45],"must":[46],"repeatedly":[47],"interpret":[48],"scratch":[51],"reason":[53],"over":[54],"noisy":[55],"verbose":[57],"navigable":[58,129],"candidates":[59],"at":[60,92],"each":[61],"step.":[62],"In":[63],"this":[64],"paper,":[65],"we":[66],"propose":[67],"a":[68],"retrieval-augmented":[69,205],"framework":[70],"improve":[72],"efficiency":[74],"stability":[76],"of":[77,149],"LLM-based":[78,216],"VLN":[79],"without":[80],"modifying":[81],"or":[82],"fine-tuning":[83],"underlying":[85],"model.":[87],"Our":[88],"approach":[89],"introduces":[90],"retrieval":[91,141,187],"two":[93],"complementary":[94,192],"levels.":[95],"At":[96,119],"episode":[98],"level,":[99,122],"instruction-level":[101,185],"embedding":[102],"retriever":[103,126],"selects":[104],"semantically":[105],"similar":[106],"successful":[107],"trajectories":[109],"in-context":[111],"exemplars,":[112],"providing":[113],"task-specific":[114],"priors":[115],"for":[116,214],"instruction":[117],"grounding.":[118],"step":[121],"imitation-learned":[124],"candidate":[125,189],"prunes":[127],"irrelevant":[128],"directions":[130],"before":[131],"inference,":[133],"reducing":[134],"action":[135],"ambiguity":[136],"prompt":[138],"complexity.":[139],"Both":[140],"modules":[142],"are":[143],"lightweight,":[144],"modular,":[145],"trained":[147],"independently":[148],"LLM.":[151],"We":[152],"evaluate":[153],"our":[154],"method":[155],"on":[156,174],"Room-to-Room":[158],"(R2R)":[159],"benchmark.":[160],"Experimental":[161],"results":[162,202],"demonstrate":[163],"consistent":[164],"improvements":[165],"in":[166],"Success":[167,170],"Rate,":[168,171],"Oracle":[169],"SPL":[173],"both":[175],"seen":[176],"Ablation":[180],"studies":[181],"further":[182],"show":[183],"that":[184,204],"exemplar":[186],"pruning":[190],"contribute":[191],"benefits":[193],"global":[195],"guidance":[196],"step-wise":[198],"decision":[199,206],"efficiency.":[200],"These":[201],"indicate":[203],"support":[207],"is":[208],"effective":[210],"scalable":[212],"strategy":[213],"enhancing":[215],"vision-and-language":[217],"navigation.":[218]},"counts_by_year":[],"updated_date":"2026-02-19T06:31:58.851227","created_date":"2026-02-19T00:00:00"}