{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JAKPY3YFRQUHTX2ZU7AIMJAJVF","short_pith_number":"pith:JAKPY3YF","schema_version":"1.0","canonical_sha256":"4814fc6f058c2879df59a7c0862409a9590bec78810d949a4b552d9249e00903","source":{"kind":"arxiv","id":"2508.10416","version":1},"attestation_state":"computed","paper":{"title":"CorrectNav: Self-Correction Flywheel Empowers Vision-Language-Action Navigation Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV"],"primary_cat":"cs.RO","authors_text":"Chengyan Zeng, Hao Dong, Hongwei Fan, Jiyao Zhang, Yuxing Long, Zhuoyuan Yu, Zihan Yang","submitted_at":"2025-08-14T07:39:26Z","abstract_excerpt":"Existing vision-and-language navigation models often deviate from the correct trajectory when executing instructions. However, these models lack effective error correction capability, hindering their recovery from errors. To address this challenge, we propose Self-correction Flywheel, a novel post-training paradigm. Instead of considering the model's error trajectories on the training set as a drawback, our paradigm emphasizes their significance as a valuable data source. We have developed a method to identify deviations in these error trajectories and devised innovative techniques to automati"},"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":"2508.10416","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2025-08-14T07:39:26Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CV"],"title_canon_sha256":"0a3f9d7d77e7640d3f70d223c6e3b273f6b4a6cebda78c67ea131133f0ecd60f","abstract_canon_sha256":"f368d5c29cf90c49c7ce353584e7be657df58eef0490da64913f5a4224e1d863"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:55.826803Z","signature_b64":"acAfERUpV0+Cw7pgCSnKCRZQkN+BOe7MR0eV9lwMNb91VXWB9aS1QMYlks6J6z7VqOuh4qN4xFbLXNF4T+FcAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4814fc6f058c2879df59a7c0862409a9590bec78810d949a4b552d9249e00903","last_reissued_at":"2026-07-05T11:53:55.826362Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:55.826362Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CorrectNav: Self-Correction Flywheel Empowers Vision-Language-Action Navigation Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV"],"primary_cat":"cs.RO","authors_text":"Chengyan Zeng, Hao Dong, Hongwei Fan, Jiyao Zhang, Yuxing Long, Zhuoyuan Yu, Zihan Yang","submitted_at":"2025-08-14T07:39:26Z","abstract_excerpt":"Existing vision-and-language navigation models often deviate from the correct trajectory when executing instructions. However, these models lack effective error correction capability, hindering their recovery from errors. To address this challenge, we propose Self-correction Flywheel, a novel post-training paradigm. Instead of considering the model's error trajectories on the training set as a drawback, our paradigm emphasizes their significance as a valuable data source. We have developed a method to identify deviations in these error trajectories and devised innovative techniques to automati"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.10416","kind":"arxiv","version":1},"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/2508.10416/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":"2508.10416","created_at":"2026-07-05T11:53:55.826421+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.10416v1","created_at":"2026-07-05T11:53:55.826421+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.10416","created_at":"2026-07-05T11:53:55.826421+00:00"},{"alias_kind":"pith_short_12","alias_value":"JAKPY3YFRQUH","created_at":"2026-07-05T11:53:55.826421+00:00"},{"alias_kind":"pith_short_16","alias_value":"JAKPY3YFRQUHTX2Z","created_at":"2026-07-05T11:53:55.826421+00:00"},{"alias_kind":"pith_short_8","alias_value":"JAKPY3YF","created_at":"2026-07-05T11:53:55.826421+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27807","citing_title":"SpikeVLA: Vision-Language-Action Models with Spiking Neural Networks","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22816","citing_title":"AwareVLN: Reasoning with Self-awareness for Vision-Language Navigation","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2512.21714","citing_title":"AstraNav-World: World Model for Foresight Control and Consistency","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13328","citing_title":"What Limits Vision-and-Language Navigation ?","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07381","citing_title":"Escaping the Diversity Trap in Robotic Manipulation via Anchor-Centric Adaptation","ref_index":74,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17473","citing_title":"Dual-Anchoring: Addressing State Drift in Vision-Language Navigation","ref_index":148,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JAKPY3YFRQUHTX2ZU7AIMJAJVF","json":"https://pith.science/pith/JAKPY3YFRQUHTX2ZU7AIMJAJVF.json","graph_json":"https://pith.science/api/pith-number/JAKPY3YFRQUHTX2ZU7AIMJAJVF/graph.json","events_json":"https://pith.science/api/pith-number/JAKPY3YFRQUHTX2ZU7AIMJAJVF/events.json","paper":"https://pith.science/paper/JAKPY3YF"},"agent_actions":{"view_html":"https://pith.science/pith/JAKPY3YFRQUHTX2ZU7AIMJAJVF","download_json":"https://pith.science/pith/JAKPY3YFRQUHTX2ZU7AIMJAJVF.json","view_paper":"https://pith.science/paper/JAKPY3YF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.10416&json=true","fetch_graph":"https://pith.science/api/pith-number/JAKPY3YFRQUHTX2ZU7AIMJAJVF/graph.json","fetch_events":"https://pith.science/api/pith-number/JAKPY3YFRQUHTX2ZU7AIMJAJVF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JAKPY3YFRQUHTX2ZU7AIMJAJVF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JAKPY3YFRQUHTX2ZU7AIMJAJVF/action/storage_attestation","attest_author":"https://pith.science/pith/JAKPY3YFRQUHTX2ZU7AIMJAJVF/action/author_attestation","sign_citation":"https://pith.science/pith/JAKPY3YFRQUHTX2ZU7AIMJAJVF/action/citation_signature","submit_replication":"https://pith.science/pith/JAKPY3YFRQUHTX2ZU7AIMJAJVF/action/replication_record"}},"created_at":"2026-07-05T11:53:55.826421+00:00","updated_at":"2026-07-05T11:53:55.826421+00:00"}