{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:LTG7HVJNM7BFKHGGVGJV7DDRWY","short_pith_number":"pith:LTG7HVJN","schema_version":"1.0","canonical_sha256":"5ccdf3d52d67c2551cc6a9935f8c71b6293cf591d2dea225968dc56b48aec784","source":{"kind":"arxiv","id":"2506.17221","version":2},"attestation_state":"computed","paper":{"title":"VLN-R1: Vision-Language Navigation via Reinforcement Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hengshuang Zhao, Jiaqi Wang, Yizhou Yu, Zhangyang Qi, Zhixiong Zhang","submitted_at":"2025-06-20T17:59:59Z","abstract_excerpt":"Vision-Language Navigation (VLN) is a core challenge in embodied AI, requiring agents to navigate real-world environments using natural language instructions. Current language model-based navigation systems operate on discrete topological graphs, limiting path planning to predefined node connections. We propose VLN-R1, an end-to-end framework that leverages Large Vision-Language Models (LVLM) to directly translate egocentric video streams into continuous navigation actions, adopting GRPO-based training inspired by DeepSeek-R1. To enable effective training, we first construct the VLN-Ego datase"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2506.17221","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-20T17:59:59Z","cross_cats_sorted":[],"title_canon_sha256":"6e852310a52152ef282e0695f2f89fae75be489b6b5def1188a21a871675c76e","abstract_canon_sha256":"9cc353972e1a83eac0440a1c1dc00302f939a0c8b4d05ac6bbf7e7e91fcbbaba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:26:53.241605Z","signature_b64":"1nzTP9hhCr/1fP0HIWmgOFWbt9Z2/K+KkI7RQ6T2nijAFEg0GRPyO+iyFX2iYHQwtZxFzHdVKXPPuoYO9b83Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5ccdf3d52d67c2551cc6a9935f8c71b6293cf591d2dea225968dc56b48aec784","last_reissued_at":"2026-07-05T11:26:53.241113Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:26:53.241113Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VLN-R1: Vision-Language Navigation via Reinforcement Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hengshuang Zhao, Jiaqi Wang, Yizhou Yu, Zhangyang Qi, Zhixiong Zhang","submitted_at":"2025-06-20T17:59:59Z","abstract_excerpt":"Vision-Language Navigation (VLN) is a core challenge in embodied AI, requiring agents to navigate real-world environments using natural language instructions. Current language model-based navigation systems operate on discrete topological graphs, limiting path planning to predefined node connections. We propose VLN-R1, an end-to-end framework that leverages Large Vision-Language Models (LVLM) to directly translate egocentric video streams into continuous navigation actions, adopting GRPO-based training inspired by DeepSeek-R1. To enable effective training, we first construct the VLN-Ego datase"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17221","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2506.17221/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2506.17221","created_at":"2026-07-05T11:26:53.241173+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.17221v2","created_at":"2026-07-05T11:26:53.241173+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17221","created_at":"2026-07-05T11:26:53.241173+00:00"},{"alias_kind":"pith_short_12","alias_value":"LTG7HVJNM7BF","created_at":"2026-07-05T11:26:53.241173+00:00"},{"alias_kind":"pith_short_16","alias_value":"LTG7HVJNM7BFKHGG","created_at":"2026-07-05T11:26:53.241173+00:00"},{"alias_kind":"pith_short_8","alias_value":"LTG7HVJN","created_at":"2026-07-05T11:26:53.241173+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":20,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.18483","citing_title":"Steadily moving semi-infinite fracture in plane poroelasticity","ref_index":77,"is_internal_anchor":true},{"citing_arxiv_id":"2607.01754","citing_title":"Path-level Hindsight Instructions for Semantic Exploration in Vision-Language Navigation","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07244","citing_title":"Beyond Waypoints: A Trajectory-Centric Waypointing Paradigm for Vision-Language Navigation","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07433","citing_title":"Watch, Remember, Reason: Human-View Video Understanding with MLLMs","ref_index":239,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01621","citing_title":"Goal2Pixel: Grounding Goals to Pixels for Vision-Language Navigation","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00963","citing_title":"Reasmory: 3D Reconstruction as Explicit Memory for VLMs Spatial Reasoning","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24578","citing_title":"World Models as Group Actions","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00095","citing_title":"Bridging the 2D-3D Gap: A Hierarchical Semantic-Geometric Map for Vision Language Navigation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22816","citing_title":"AwareVLN: Reasoning with Self-awareness for Vision-Language Navigation","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15964","citing_title":"WorldVLN: Autoregressive World Action Model for Aerial Vision-Language Navigation","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2512.08639","citing_title":"Aerial Vision-Language Navigation with a Unified Framework for Spatial, Temporal and Embodied Reasoning","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02870","citing_title":"Token Warping Helps MLLMs Look from Nearby Viewpoints","ref_index":76,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04108","citing_title":"Hypothesis Graph Refinement: Hypothesis-Driven Exploration with Cascade Error Correction for Embodied Navigation","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27620","citing_title":"SpaAct: Spatially-Activated Transition Learning with Curriculum Adaptation for Vision-Language Navigation","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09146","citing_title":"Beyond Thinking: Imagining in 360$^\\circ$ for Humanoid Visual Search","ref_index":84,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08232","citing_title":"HiRO-Nav: Hybrid ReasOning Enables Efficient Embodied Navigation","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06725","citing_title":"Enhancing MLLM Spatial Understanding via Active 3D Scene Exploration for Multi-Perspective Reasoning","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17473","citing_title":"Dual-Anchoring: Addressing State Drift in Vision-Language Navigation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17407","citing_title":"Think before Go: Hierarchical Reasoning for Image-goal Navigation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18484","citing_title":"XEmbodied: A Foundation Model with Enhanced Geometric and Physical Cues for Large-Scale Embodied Environments","ref_index":77,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LTG7HVJNM7BFKHGGVGJV7DDRWY","json":"https://pith.science/pith/LTG7HVJNM7BFKHGGVGJV7DDRWY.json","graph_json":"https://pith.science/api/pith-number/LTG7HVJNM7BFKHGGVGJV7DDRWY/graph.json","events_json":"https://pith.science/api/pith-number/LTG7HVJNM7BFKHGGVGJV7DDRWY/events.json","paper":"https://pith.science/paper/LTG7HVJN"},"agent_actions":{"view_html":"https://pith.science/pith/LTG7HVJNM7BFKHGGVGJV7DDRWY","download_json":"https://pith.science/pith/LTG7HVJNM7BFKHGGVGJV7DDRWY.json","view_paper":"https://pith.science/paper/LTG7HVJN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.17221&json=true","fetch_graph":"https://pith.science/api/pith-number/LTG7HVJNM7BFKHGGVGJV7DDRWY/graph.json","fetch_events":"https://pith.science/api/pith-number/LTG7HVJNM7BFKHGGVGJV7DDRWY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LTG7HVJNM7BFKHGGVGJV7DDRWY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LTG7HVJNM7BFKHGGVGJV7DDRWY/action/storage_attestation","attest_author":"https://pith.science/pith/LTG7HVJNM7BFKHGGVGJV7DDRWY/action/author_attestation","sign_citation":"https://pith.science/pith/LTG7HVJNM7BFKHGGVGJV7DDRWY/action/citation_signature","submit_replication":"https://pith.science/pith/LTG7HVJNM7BFKHGGVGJV7DDRWY/action/replication_record"}},"created_at":"2026-07-05T11:26:53.241173+00:00","updated_at":"2026-07-05T11:26:53.241173+00:00"}