{"id":"https://openalex.org/W7154401673","doi":"https://doi.org/10.48550/arxiv.2604.09604","title":"LLMs for Text-Based Exploration and Navigation Under Partial Observability","display_name":"LLMs for Text-Based Exploration and Navigation Under Partial Observability","publication_year":2026,"publication_date":"2026-03-10","ids":{"openalex":"https://openalex.org/W7154401673","doi":"https://doi.org/10.48550/arxiv.2604.09604"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.09604","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09604","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.2604.09604","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5048143522","display_name":"Stephan Sandfuchs","orcid":"https://orcid.org/0000-0001-7621-7191"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Sandfuchs, Stephan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133583001","display_name":"Maximilian Melchert","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Melchert, Maximilian","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5027576910","display_name":"J\u00f6rg Frochte","orcid":"https://orcid.org/0000-0002-5908-5649"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Frochte, J\u00f6rg","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5048143522"],"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":0.4171000123023987,"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":0.4171000123023987,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.13230000436306,"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/T10586","display_name":"Robotic Path Planning Algorithms","score":0.05990000069141388,"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/observability","display_name":"Observability","score":0.7888000011444092},{"id":"https://openalex.org/keywords/oracle","display_name":"Oracle","score":0.7311999797821045},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5961999893188477},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.4830000102519989},{"id":"https://openalex.org/keywords/correctness","display_name":"Correctness","score":0.48249998688697815},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.44839999079704285},{"id":"https://openalex.org/keywords/function","display_name":"Function (biology)","score":0.44429999589920044}],"concepts":[{"id":"https://openalex.org/C36299963","wikidata":"https://www.wikidata.org/wiki/Q1369844","display_name":"Observability","level":2,"score":0.7888000011444092},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.7311999797821045},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6402999758720398},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5961999893188477},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.4830000102519989},{"id":"https://openalex.org/C55439883","wikidata":"https://www.wikidata.org/wiki/Q360812","display_name":"Correctness","level":2,"score":0.48249998688697815},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.44839999079704285},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.44429999589920044},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3779999911785126},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35850000381469727},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.34040001034736633},{"id":"https://openalex.org/C2780735816","wikidata":"https://www.wikidata.org/wiki/Q28324931","display_name":"Incremental learning","level":2,"score":0.335099995136261},{"id":"https://openalex.org/C198647972","wikidata":"https://www.wikidata.org/wiki/Q44475","display_name":"Spear","level":2,"score":0.3246000111103058},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3228999972343445},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3149999976158142},{"id":"https://openalex.org/C183115368","wikidata":"https://www.wikidata.org/wiki/Q856577","display_name":"Weighting","level":2,"score":0.31349998712539673},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2906999886035919},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.28870001435279846},{"id":"https://openalex.org/C115051666","wikidata":"https://www.wikidata.org/wiki/Q6522493","display_name":"Ranging","level":2,"score":0.274399995803833},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.273499995470047},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C42475967","wikidata":"https://www.wikidata.org/wiki/Q194292","display_name":"Operations research","level":1,"score":0.2727999985218048},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2635999917984009},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2635999917984009}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.09604","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09604","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.2604.09604","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09604","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Exploration":[0],"and":[1,12,60,79,98,123,162,188],"goal-directed":[2],"navigation":[3,137],"in":[4,45,150],"unknown":[5],"layouts":[6,90],"are":[7,83,110],"central":[8],"to":[9,212],"inspection,":[10],"logistics,":[11],"search-and-rescue.":[13],"We":[14,37,172],"ask":[15],"whether":[16],"large":[17],"language":[18],"models":[19,134,157,169],"(LLMs)":[20],"can":[21,179],"function":[22],"as":[23,120,128,130,208],"\\emph{text-only}":[24],"controllers":[25],"under":[26,182],"partial":[27,183,214],"observability":[28],"--":[29],"without":[30],"code":[31],"execution,":[32],"tools,":[33],"or":[34],"program":[35],"synthesis.":[36],"introduce":[38],"a":[39,53,209],"reproducible":[40],"benchmark":[41],"with":[42,204],"oracle":[43,127,146],"localisation":[44],"fixed":[46],"ASCII":[47],"gridworlds:":[48],"each":[49],"step":[50],"reveals":[51],"only":[52],"local":[54],"$5\\times5$":[55],"window":[56],"around":[57],"the":[58,61,101,104,151],"agent":[59],"model":[62],"must":[63],"select":[64],"one":[65],"of":[66,77,91],"\\texttt{UP/RIGHT/DOWN/LEFT}.":[67],"Nine":[68],"contemporary":[69],"LLMs":[70],"ranging":[71],"from":[72],"open/proprietary,":[73],"dense":[74,167],"/":[75],"Mixture":[76],"Experts":[78],"instruction-":[80],"vs.":[81,126],"reasoning-tuned":[82],"evaluated":[84,111],"on":[85,103,112],"two":[86],"tasks":[87],"across":[88,138],"three":[89],"increasing":[92],"difficulty:":[93],"\\emph{Exploration}":[94],"(maximising":[95],"revealed":[96],"cells)":[97],"\\emph{Navigation}":[99],"(reach":[100],"goal":[102],"shortest":[105],"path).":[106],"The":[107],"experimental":[108],"results":[109],"quantitative":[113],"metrics":[114],"including":[115],"\\emph{success":[116],"rate},":[117],"\\emph{efficiency}":[118],"such":[119],"normalised":[121],"coverage":[122],"\\emph{path":[124],"length}":[125],"well":[129],"qualitative":[131],"analysis.":[132],"Reasoning-tuned":[133,156],"reliably":[135],"complete":[136],"all":[139],"layouts,":[140],"yet":[141],"remain":[142,170],"less":[143],"efficient":[144],"than":[145,195],"paths.":[147],"Few-shot":[148],"demonstrations":[149],"prompt":[152],"chiefly":[153],"help":[154],"these":[155],"by":[158],"reducing":[159],"invalid":[160],"moves":[161],"shortening":[163],"paths,":[164],"while":[165],"classic":[166],"instruction":[168],"inconsistent.":[171],"observe":[173],"characteristic":[174],"action":[175],"priors":[176],"(UP/RIGHT)":[177],"that":[178],"induce":[180],"looping":[181],"observability.":[184],"Overall,":[185],"training":[186],"regimen":[187],"test-time":[189],"deliberation":[190],"predict":[191],"control":[192],"ability":[193],"better":[194],"raw":[196],"parameter":[197],"count.":[198],"These":[199],"findings":[200],"suggest":[201],"lightweight":[202],"hybridisation":[203],"classical":[205],"online":[206],"planners":[207],"practical":[210],"route":[211],"deployable":[213],"map":[215],"systems.":[216]},"counts_by_year":[],"updated_date":"2026-04-15T06:04:33.058270","created_date":"2026-04-15T00:00:00"}