{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VL3QBDP6JP4DJTHRN3GOY5JRZH","short_pith_number":"pith:VL3QBDP6","schema_version":"1.0","canonical_sha256":"aaf7008dfe4bf834ccf16eccec7531c9c8d8234367aac13a017f3bbd097619f6","source":{"kind":"arxiv","id":"2607.01586","version":1},"attestation_state":"computed","paper":{"title":"VLAFlow: A Unified Training Framework for Vision-Language-Action Models via Co-training and Future Latent Alignment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.CV","authors_text":"Fangxiang Feng, Fengfa Li, Guoyang Xia, Hongjin Ji, Kun Zhan, Lei Ren, Yan Xie","submitted_at":"2026-07-02T01:38:16Z","abstract_excerpt":"Vision-language-action models (VLAs) have recently advanced robotic manipulation, yet the effects of different robot-data pre-training paradigms remain difficult to compare because existing models often differ in architecture, data, action space, and evaluation protocol. We present VLAFlow (Vision-Language-Action Flow), a unified flow-matching framework for controlled comparison of VLA training objectives. Using a heterogeneous robot corpus, OXEMix, containing approximately 5,000 hours of data from DROID, OpenX-Embodiment, OpenX-Augmented, and RoboCOIN, we evaluate four paradigms under the sam"},"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":"2607.01586","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-07-02T01:38:16Z","cross_cats_sorted":["cs.AI","cs.RO"],"title_canon_sha256":"4e7c8af2a0bec66d498230c6b88b39b56243ef951f58f3eff908d53bdc54f04f","abstract_canon_sha256":"e94a61db5f9c1626f3953a91a88e64b60b0b54061aac042264738d24e25fe805"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:02.643040Z","signature_b64":"FNl81sWSh4Pg8C2h3mIfojN2OMH/sdx9Zk4whI3QBwuoPF02XGix8camrYHN21sXf8rQkWVgN2PrgFNlPbH+Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aaf7008dfe4bf834ccf16eccec7531c9c8d8234367aac13a017f3bbd097619f6","last_reissued_at":"2026-07-03T01:17:02.642172Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:02.642172Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VLAFlow: A Unified Training Framework for Vision-Language-Action Models via Co-training and Future Latent Alignment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.RO"],"primary_cat":"cs.CV","authors_text":"Fangxiang Feng, Fengfa Li, Guoyang Xia, Hongjin Ji, Kun Zhan, Lei Ren, Yan Xie","submitted_at":"2026-07-02T01:38:16Z","abstract_excerpt":"Vision-language-action models (VLAs) have recently advanced robotic manipulation, yet the effects of different robot-data pre-training paradigms remain difficult to compare because existing models often differ in architecture, data, action space, and evaluation protocol. We present VLAFlow (Vision-Language-Action Flow), a unified flow-matching framework for controlled comparison of VLA training objectives. Using a heterogeneous robot corpus, OXEMix, containing approximately 5,000 hours of data from DROID, OpenX-Embodiment, OpenX-Augmented, and RoboCOIN, we evaluate four paradigms under the sam"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.01586","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/2607.01586/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":"2607.01586","created_at":"2026-07-03T01:17:02.642230+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.01586v1","created_at":"2026-07-03T01:17:02.642230+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.01586","created_at":"2026-07-03T01:17:02.642230+00:00"},{"alias_kind":"pith_short_12","alias_value":"VL3QBDP6JP4D","created_at":"2026-07-03T01:17:02.642230+00:00"},{"alias_kind":"pith_short_16","alias_value":"VL3QBDP6JP4DJTHR","created_at":"2026-07-03T01:17:02.642230+00:00"},{"alias_kind":"pith_short_8","alias_value":"VL3QBDP6","created_at":"2026-07-03T01:17:02.642230+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VL3QBDP6JP4DJTHRN3GOY5JRZH","json":"https://pith.science/pith/VL3QBDP6JP4DJTHRN3GOY5JRZH.json","graph_json":"https://pith.science/api/pith-number/VL3QBDP6JP4DJTHRN3GOY5JRZH/graph.json","events_json":"https://pith.science/api/pith-number/VL3QBDP6JP4DJTHRN3GOY5JRZH/events.json","paper":"https://pith.science/paper/VL3QBDP6"},"agent_actions":{"view_html":"https://pith.science/pith/VL3QBDP6JP4DJTHRN3GOY5JRZH","download_json":"https://pith.science/pith/VL3QBDP6JP4DJTHRN3GOY5JRZH.json","view_paper":"https://pith.science/paper/VL3QBDP6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.01586&json=true","fetch_graph":"https://pith.science/api/pith-number/VL3QBDP6JP4DJTHRN3GOY5JRZH/graph.json","fetch_events":"https://pith.science/api/pith-number/VL3QBDP6JP4DJTHRN3GOY5JRZH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VL3QBDP6JP4DJTHRN3GOY5JRZH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VL3QBDP6JP4DJTHRN3GOY5JRZH/action/storage_attestation","attest_author":"https://pith.science/pith/VL3QBDP6JP4DJTHRN3GOY5JRZH/action/author_attestation","sign_citation":"https://pith.science/pith/VL3QBDP6JP4DJTHRN3GOY5JRZH/action/citation_signature","submit_replication":"https://pith.science/pith/VL3QBDP6JP4DJTHRN3GOY5JRZH/action/replication_record"}},"created_at":"2026-07-03T01:17:02.642230+00:00","updated_at":"2026-07-03T01:17:02.642230+00:00"}