{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TXJR54DUZMEVE4AZH2Q5WVU6IU","short_pith_number":"pith:TXJR54DU","schema_version":"1.0","canonical_sha256":"9dd31ef074cb095270193ea1db569e450a6bd041d883ec4a76d44b21e9deffe5","source":{"kind":"arxiv","id":"2602.06508","version":2},"attestation_state":"computed","paper":{"title":"World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Hai Ci, Kevin Yuchen Ma, Mike Zheng Shou, Xiaokang Liu, Zechen Bai","submitted_at":"2026-02-06T08:57:55Z","abstract_excerpt":"Reinforcement learning (RL) can refine Vision-Language-Action (VLA) policies beyond behavior cloning, but real-world RL remains expensive due to extensive rollouts, resets, supervision, and safety risks. Action-conditioned video world models offer an option to train in virtual environments, yet they exhibit imprecise action following, particularly on subtle near-success failures. Besides, they lack native reward signals for RL. Computing rewards based on inaccurate visual predictions remain unreliable. We introduce World-VLA-Loop, structured around two foundational designs and a higher-level c"},"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":"2602.06508","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-02-06T08:57:55Z","cross_cats_sorted":[],"title_canon_sha256":"15b9f478587eada233616d273554b93b613d0cb6d6dc6a71d55a9566c65bf097","abstract_canon_sha256":"435fc8d99593b354349fae080aed462e4d1c4e3df96863e2e1e0bd819b9e0bde"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:04:05.196997Z","signature_b64":"GzRve1GWzckqbZHdyID2M9lSyx3bzDGVe+tRLrrgKeRPfr8G9kikAy1VBmf+7v8iaj7/Vr02O7nrBL/DZJVtBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9dd31ef074cb095270193ea1db569e450a6bd041d883ec4a76d44b21e9deffe5","last_reissued_at":"2026-05-26T02:04:05.196105Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:04:05.196105Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"World-VLA-Loop: Closed-Loop Learning of Video World Model and VLA Policy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Hai Ci, Kevin Yuchen Ma, Mike Zheng Shou, Xiaokang Liu, Zechen Bai","submitted_at":"2026-02-06T08:57:55Z","abstract_excerpt":"Reinforcement learning (RL) can refine Vision-Language-Action (VLA) policies beyond behavior cloning, but real-world RL remains expensive due to extensive rollouts, resets, supervision, and safety risks. Action-conditioned video world models offer an option to train in virtual environments, yet they exhibit imprecise action following, particularly on subtle near-success failures. Besides, they lack native reward signals for RL. Computing rewards based on inaccurate visual predictions remain unreliable. We introduce World-VLA-Loop, structured around two foundational designs and a higher-level c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.06508","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/2602.06508/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":"2602.06508","created_at":"2026-05-26T02:04:05.196248+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.06508v2","created_at":"2026-05-26T02:04:05.196248+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.06508","created_at":"2026-05-26T02:04:05.196248+00:00"},{"alias_kind":"pith_short_12","alias_value":"TXJR54DUZMEV","created_at":"2026-05-26T02:04:05.196248+00:00"},{"alias_kind":"pith_short_16","alias_value":"TXJR54DUZMEVE4AZ","created_at":"2026-05-26T02:04:05.196248+00:00"},{"alias_kind":"pith_short_8","alias_value":"TXJR54DU","created_at":"2026-05-26T02:04:05.196248+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":11,"sample":[{"citing_arxiv_id":"2605.22446","citing_title":"Pre-VLA: Preemptive Runtime Verification for Reliable Vision-Language-Action and World-Model Rollouts","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12334","citing_title":"Reinforcing VLAs in Task-Agnostic World Models","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16154","citing_title":"Learn Where Outcomes Diverge: Efficient VLA RL via Probabilistic Chunk Masking","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15705","citing_title":"Feedback World Model Enables Precise Guidance of Diffusion Policy","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09423","citing_title":"SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learning","ref_index":50,"is_internal_anchor":true},{"citing_arxiv_id":"2604.02965","citing_title":"Open-Loop Planning, Closed-Loop Verification: Speculative Verification for VLA","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12334","citing_title":"Reinforcing VLAs in Task-Agnostic World Models","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2604.27711","citing_title":"ExoActor: Exocentric Video Generation as Generalizable Interactive Humanoid Control","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09423","citing_title":"SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learning","ref_index":50,"is_internal_anchor":true},{"citing_arxiv_id":"2605.00080","citing_title":"World Model for Robot Learning: A Comprehensive Survey","ref_index":34,"is_internal_anchor":true},{"citing_arxiv_id":"2604.21741","citing_title":"Hi-WM: Human-in-the-World-Model for Scalable Robot Post-Training","ref_index":38,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TXJR54DUZMEVE4AZH2Q5WVU6IU","json":"https://pith.science/pith/TXJR54DUZMEVE4AZH2Q5WVU6IU.json","graph_json":"https://pith.science/api/pith-number/TXJR54DUZMEVE4AZH2Q5WVU6IU/graph.json","events_json":"https://pith.science/api/pith-number/TXJR54DUZMEVE4AZH2Q5WVU6IU/events.json","paper":"https://pith.science/paper/TXJR54DU"},"agent_actions":{"view_html":"https://pith.science/pith/TXJR54DUZMEVE4AZH2Q5WVU6IU","download_json":"https://pith.science/pith/TXJR54DUZMEVE4AZH2Q5WVU6IU.json","view_paper":"https://pith.science/paper/TXJR54DU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.06508&json=true","fetch_graph":"https://pith.science/api/pith-number/TXJR54DUZMEVE4AZH2Q5WVU6IU/graph.json","fetch_events":"https://pith.science/api/pith-number/TXJR54DUZMEVE4AZH2Q5WVU6IU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TXJR54DUZMEVE4AZH2Q5WVU6IU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TXJR54DUZMEVE4AZH2Q5WVU6IU/action/storage_attestation","attest_author":"https://pith.science/pith/TXJR54DUZMEVE4AZH2Q5WVU6IU/action/author_attestation","sign_citation":"https://pith.science/pith/TXJR54DUZMEVE4AZH2Q5WVU6IU/action/citation_signature","submit_replication":"https://pith.science/pith/TXJR54DUZMEVE4AZH2Q5WVU6IU/action/replication_record"}},"created_at":"2026-05-26T02:04:05.196248+00:00","updated_at":"2026-05-26T02:04:05.196248+00:00"}