{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:WARCALHMZSQRAVAFK4UNG4GN6Z","short_pith_number":"pith:WARCALHM","schema_version":"1.0","canonical_sha256":"b022202ceccca11054055728d370cdf6623787a2d9f725cc986a4f0b6a8bd54e","source":{"kind":"arxiv","id":"2603.10422","version":2},"attestation_state":"computed","paper":{"title":"World2Act: Latent Action Post-Training from World Model Dynamics","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Abdullah Sohail, An Dinh Vuong, Anqing Duan, Haoran Ding, Ian Reid, Ivan Laptev, Liang Ma, Tuan Van Vo, Xiaodan Liang","submitted_at":"2026-03-11T05:11:44Z","abstract_excerpt":"World Models (WMs) offer a promising mechanism for post-training Vision-Language-Action (VLA) policies by providing dynamics priors that improve generalization under task and scene variation. However, most WM-based post-training methods rely on pixel-space supervision, making policies sensitive to visual artifacts introduced by imperfect WM rollouts. We present World2Act, a latent-space post-training framework that transfers WM dynamics to the VLA policy without pixel-space supervision. World2Act operates in two stages: 1) it induces a shared video-action latent space by contrastively aligning"},"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":"2603.10422","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-03-11T05:11:44Z","cross_cats_sorted":[],"title_canon_sha256":"e74966c3ef677d99cf0adbca73c5e420cf5f350bcc66f040cd8ddfdc9c92e4b9","abstract_canon_sha256":"e0923c4284e7b8fed756513c19bd505f5fbb83c28a75c145f7b64f6d416fa51f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:02:36.833769Z","signature_b64":"E5s50dbgK0cfi6ZA2i/imz8XODuIsVs19J8p8I9DrldozOcBh3BBD+M+zLduoLk5JM7vQ20IXBXNuEGQl8/2BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b022202ceccca11054055728d370cdf6623787a2d9f725cc986a4f0b6a8bd54e","last_reissued_at":"2026-06-01T01:02:36.832665Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:02:36.832665Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"World2Act: Latent Action Post-Training from World Model Dynamics","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Abdullah Sohail, An Dinh Vuong, Anqing Duan, Haoran Ding, Ian Reid, Ivan Laptev, Liang Ma, Tuan Van Vo, Xiaodan Liang","submitted_at":"2026-03-11T05:11:44Z","abstract_excerpt":"World Models (WMs) offer a promising mechanism for post-training Vision-Language-Action (VLA) policies by providing dynamics priors that improve generalization under task and scene variation. However, most WM-based post-training methods rely on pixel-space supervision, making policies sensitive to visual artifacts introduced by imperfect WM rollouts. We present World2Act, a latent-space post-training framework that transfers WM dynamics to the VLA policy without pixel-space supervision. World2Act operates in two stages: 1) it induces a shared video-action latent space by contrastively aligning"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.10422","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/2603.10422/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":"2603.10422","created_at":"2026-06-01T01:02:36.832816+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.10422v2","created_at":"2026-06-01T01:02:36.832816+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.10422","created_at":"2026-06-01T01:02:36.832816+00:00"},{"alias_kind":"pith_short_12","alias_value":"WARCALHMZSQR","created_at":"2026-06-01T01:02:36.832816+00:00"},{"alias_kind":"pith_short_16","alias_value":"WARCALHMZSQRAVAF","created_at":"2026-06-01T01:02:36.832816+00:00"},{"alias_kind":"pith_short_8","alias_value":"WARCALHM","created_at":"2026-06-01T01:02:36.832816+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/WARCALHMZSQRAVAFK4UNG4GN6Z","json":"https://pith.science/pith/WARCALHMZSQRAVAFK4UNG4GN6Z.json","graph_json":"https://pith.science/api/pith-number/WARCALHMZSQRAVAFK4UNG4GN6Z/graph.json","events_json":"https://pith.science/api/pith-number/WARCALHMZSQRAVAFK4UNG4GN6Z/events.json","paper":"https://pith.science/paper/WARCALHM"},"agent_actions":{"view_html":"https://pith.science/pith/WARCALHMZSQRAVAFK4UNG4GN6Z","download_json":"https://pith.science/pith/WARCALHMZSQRAVAFK4UNG4GN6Z.json","view_paper":"https://pith.science/paper/WARCALHM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.10422&json=true","fetch_graph":"https://pith.science/api/pith-number/WARCALHMZSQRAVAFK4UNG4GN6Z/graph.json","fetch_events":"https://pith.science/api/pith-number/WARCALHMZSQRAVAFK4UNG4GN6Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WARCALHMZSQRAVAFK4UNG4GN6Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WARCALHMZSQRAVAFK4UNG4GN6Z/action/storage_attestation","attest_author":"https://pith.science/pith/WARCALHMZSQRAVAFK4UNG4GN6Z/action/author_attestation","sign_citation":"https://pith.science/pith/WARCALHMZSQRAVAFK4UNG4GN6Z/action/citation_signature","submit_replication":"https://pith.science/pith/WARCALHMZSQRAVAFK4UNG4GN6Z/action/replication_record"}},"created_at":"2026-06-01T01:02:36.832816+00:00","updated_at":"2026-06-01T01:02:36.832816+00:00"}