{"paper":{"title":"LMPath: Language-Mediated Priors and Path Generation for Aerial Exploration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"LMPath uses language models on satellite imagery to generate semantic priors that guide UAV search paths more efficiently than uniform geometric coverage.","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Camillo J. Taylor, Fernando Cladera, Jonathan A. Diller, Vijay Kumar","submitted_at":"2026-05-13T17:02:51Z","abstract_excerpt":"Traditional autonomous UAV search missions rely on geometric coverage patterns that ignore the semantic context of the target, leading to significant time waste in large-scale environments. In this paper we present LMPath, a pipeline for generating language-mediated exploration priors for Unmanned Aerial Vehicle (UAV) search missions that leverages semantics. Given a basic geofence and an object of interest prompt, LMPath uses generative language models to determine what regions of the environment should contain that object and a foundation vision model ran over satellite imagery to segment su"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"paths generated using LMPath outperform traditional path planning approaches for search missions","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That generative language models and foundation vision models can reliably identify regions likely to contain the prompted object from satellite imagery alone, without domain-specific fine-tuning or validation against ground truth.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"LMPath generates language-mediated priors from object prompts and satellite segmentation to produce UAV search paths that outperform traditional geometric coverage in simulations and real flights.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"LMPath uses language models on satellite imagery to generate semantic priors that guide UAV search paths more efficiently than uniform geometric coverage.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"445eda67c29d42e518bd97eab39e3257f78c3b7e9527e7dcb9c222f7510ada9c"},"source":{"id":"2605.13782","kind":"arxiv","version":1},"verdict":{"id":"757eb98d-4696-4a11-b0c0-1d83972f0709","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T17:45:02.869912Z","strongest_claim":"paths generated using LMPath outperform traditional path planning approaches for search missions","one_line_summary":"LMPath generates language-mediated priors from object prompts and satellite segmentation to produce UAV search paths that outperform traditional geometric coverage in simulations and real flights.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That generative language models and foundation vision models can reliably identify regions likely to contain the prompted object from satellite imagery alone, without domain-specific fine-tuning or validation against ground truth.","pith_extraction_headline":"LMPath uses language models on satellite imagery to generate semantic priors that guide UAV search paths more efficiently than uniform geometric coverage."},"references":{"count":10,"sample":[{"doi":"","year":2025,"title":"Proceedings of the 23rd Annual International Conference on Mobile Systems, Applications and Services , pages =","work_id":"fd60c056-52f1-461b-a118-618ec5a1d965","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"UAV-VLA: Vision-Language-Action System for Large Scale Aerial Mission Generation , year=","work_id":"72c8905e-a78d-4968-934a-f934f8cc17bb","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"HALO: High-Altitude Language-Conditioned Monocular Aerial Exploration and Navigation , author=. 2025 , eprint=","work_id":"7cded1e6-cf46-4ddd-b924-08a87f0c5bf2","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"UAV-CodeAgents: Scalable UAV Mission Planning via Multi-Agent ReAct and Vision-Language Reasoning , author=. 2025 , eprint=","work_id":"56392162-f4d9-4b6c-a97a-1c84dbc039b2","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"UAV-VLPA*: Vision-Language Guided Global-Local UAV Mission Planning from Satellite Imagery , year=","work_id":"580d82b9-c593-4170-b996-620e047f4664","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":10,"snapshot_sha256":"6120d75c7609f336cc3671e5f0c427ef662a0717e5a50604b47706d718c07623","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}