{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:M2U7PZX53CKRI4K6ZIPK3YBQ63","short_pith_number":"pith:M2U7PZX5","schema_version":"1.0","canonical_sha256":"66a9f7e6fdd89514715eca1eade030f6cb695c8bf345940335e2e2e4966cd52c","source":{"kind":"arxiv","id":"2412.08467","version":2},"attestation_state":"computed","paper":{"title":"Bootstrapping Language-Guided Navigation Learning with Self-Refining Data Flywheel","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Jialu Li, Kunchang Li, Limin Wang, Mohit Bansal, Shoubin Yu, Songze Li, Yali Wang, Yicong Hong, Yi Wang, Yu Qiao, Zun Wang","submitted_at":"2024-12-11T15:32:24Z","abstract_excerpt":"Creating high-quality data for training robust language-instructed agents is a long-lasting challenge in embodied AI. In this paper, we introduce a Self-Refining Data Flywheel (SRDF) that generates high-quality and large-scale navigational instruction-trajectory pairs by iteratively refining the data pool through the collaboration between two models, the instruction generator and the navigator, without any human-in-the-loop annotation. Specifically, SRDF starts with using a base generator to create an initial data pool for training a base navigator, followed by applying the trained navigator t"},"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":"2412.08467","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-12-11T15:32:24Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"c231c84d7406d68a4f3bdece465ce1ff4e5c9b3a613b09100d2c86afb805a057","abstract_canon_sha256":"d734895dba94110b1d6216dd9ec3730457846c64fac78d500b4638a828fbf4cd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:21:27.233727Z","signature_b64":"lg6RRKpbBgDIkphn5K//ao+dMUDBvIYG0jxFt9tFjj+hcvMTQNiuyUbOZmypeExwb1FB0yoVHdtjGQ6CrhZQCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"66a9f7e6fdd89514715eca1eade030f6cb695c8bf345940335e2e2e4966cd52c","last_reissued_at":"2026-07-05T10:21:27.233165Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:21:27.233165Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bootstrapping Language-Guided Navigation Learning with Self-Refining Data Flywheel","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Jialu Li, Kunchang Li, Limin Wang, Mohit Bansal, Shoubin Yu, Songze Li, Yali Wang, Yicong Hong, Yi Wang, Yu Qiao, Zun Wang","submitted_at":"2024-12-11T15:32:24Z","abstract_excerpt":"Creating high-quality data for training robust language-instructed agents is a long-lasting challenge in embodied AI. In this paper, we introduce a Self-Refining Data Flywheel (SRDF) that generates high-quality and large-scale navigational instruction-trajectory pairs by iteratively refining the data pool through the collaboration between two models, the instruction generator and the navigator, without any human-in-the-loop annotation. Specifically, SRDF starts with using a base generator to create an initial data pool for training a base navigator, followed by applying the trained navigator t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.08467","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/2412.08467/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":"2412.08467","created_at":"2026-07-05T10:21:27.233225+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.08467v2","created_at":"2026-07-05T10:21:27.233225+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.08467","created_at":"2026-07-05T10:21:27.233225+00:00"},{"alias_kind":"pith_short_12","alias_value":"M2U7PZX53CKR","created_at":"2026-07-05T10:21:27.233225+00:00"},{"alias_kind":"pith_short_16","alias_value":"M2U7PZX53CKRI4K6","created_at":"2026-07-05T10:21:27.233225+00:00"},{"alias_kind":"pith_short_8","alias_value":"M2U7PZX5","created_at":"2026-07-05T10:21:27.233225+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05765","citing_title":"Image2Sim: Scaling Embodied Navigation via Generative Neural Simulator","ref_index":60,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22036","citing_title":"GA-VLN: Geometry-Aware BEV Representation for Efficient Vision-Language Navigation","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2508.09547","citing_title":"GoViG: Goal-Conditioned Visual Navigation Instruction Generation via Multimodal Reasoning","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10118","citing_title":"Plan in Sandbox, Navigate in Open Worlds: Learning Physics-Grounded Abstracted Experience for Embodied Navigation","ref_index":54,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M2U7PZX53CKRI4K6ZIPK3YBQ63","json":"https://pith.science/pith/M2U7PZX53CKRI4K6ZIPK3YBQ63.json","graph_json":"https://pith.science/api/pith-number/M2U7PZX53CKRI4K6ZIPK3YBQ63/graph.json","events_json":"https://pith.science/api/pith-number/M2U7PZX53CKRI4K6ZIPK3YBQ63/events.json","paper":"https://pith.science/paper/M2U7PZX5"},"agent_actions":{"view_html":"https://pith.science/pith/M2U7PZX53CKRI4K6ZIPK3YBQ63","download_json":"https://pith.science/pith/M2U7PZX53CKRI4K6ZIPK3YBQ63.json","view_paper":"https://pith.science/paper/M2U7PZX5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.08467&json=true","fetch_graph":"https://pith.science/api/pith-number/M2U7PZX53CKRI4K6ZIPK3YBQ63/graph.json","fetch_events":"https://pith.science/api/pith-number/M2U7PZX53CKRI4K6ZIPK3YBQ63/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M2U7PZX53CKRI4K6ZIPK3YBQ63/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M2U7PZX53CKRI4K6ZIPK3YBQ63/action/storage_attestation","attest_author":"https://pith.science/pith/M2U7PZX53CKRI4K6ZIPK3YBQ63/action/author_attestation","sign_citation":"https://pith.science/pith/M2U7PZX53CKRI4K6ZIPK3YBQ63/action/citation_signature","submit_replication":"https://pith.science/pith/M2U7PZX53CKRI4K6ZIPK3YBQ63/action/replication_record"}},"created_at":"2026-07-05T10:21:27.233225+00:00","updated_at":"2026-07-05T10:21:27.233225+00:00"}