{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:6XAIUR2GYONFPTZCI4R7XQDBSB","short_pith_number":"pith:6XAIUR2G","canonical_record":{"source":{"id":"2605.14274","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T02:18:58Z","cross_cats_sorted":[],"title_canon_sha256":"1b215cebce7d0e38ab236f9d44487c275a2e490f5694c52f84d4a62268eba5b7","abstract_canon_sha256":"37ec7e054359b823a87cddfe110a26171fcb72f316bcafd8535a8a9165226c43"},"schema_version":"1.0"},"canonical_sha256":"f5c08a4746c39a57cf224723fbc061904754c270534347d31de2c1010049c53d","source":{"kind":"arxiv","id":"2605.14274","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14274","created_at":"2026-05-17T23:39:10Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14274v1","created_at":"2026-05-17T23:39:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14274","created_at":"2026-05-17T23:39:10Z"},{"alias_kind":"pith_short_12","alias_value":"6XAIUR2GYONF","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"6XAIUR2GYONFPTZC","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"6XAIUR2G","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:6XAIUR2GYONFPTZCI4R7XQDBSB","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14274","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T02:18:58Z","cross_cats_sorted":[],"title_canon_sha256":"1b215cebce7d0e38ab236f9d44487c275a2e490f5694c52f84d4a62268eba5b7","abstract_canon_sha256":"37ec7e054359b823a87cddfe110a26171fcb72f316bcafd8535a8a9165226c43"},"schema_version":"1.0"},"canonical_sha256":"f5c08a4746c39a57cf224723fbc061904754c270534347d31de2c1010049c53d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:10.362867Z","signature_b64":"1uOy3fas62FfRGgehF44zZbuFBaRK8bMArkAjSGH25mxFaxpYBzT3jJ32crAZ6qpWm0hcuG4apieGcRN+YZ1Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5c08a4746c39a57cf224723fbc061904754c270534347d31de2c1010049c53d","last_reissued_at":"2026-05-17T23:39:10.362385Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:10.362385Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14274","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"afcUQCLpkCxpDg4w7MnFLcOGprqR4yl1Kyjvf83RvMQe+ni3HMWCgTuzxaTyepIQGlY3vIGxCHJrMnucmR4tCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T20:22:43.972256Z"},"content_sha256":"bf1f782dcfde35992974afaa166c3e60b36df13cb570e273b4f9d89a4bd3ae63","schema_version":"1.0","event_id":"sha256:bf1f782dcfde35992974afaa166c3e60b36df13cb570e273b4f9d89a4bd3ae63"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:6XAIUR2GYONFPTZCI4R7XQDBSB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CreFlow: Corrective Reflow for Sparse-Reward Embodied Video Diffusion RL","license":"http://creativecommons.org/licenses/by/4.0/","headline":"CreFlow uses automatically generated Linear Temporal Logic rewards plus corrective reflow to align video diffusion rollouts with embodied task rules and lift downstream success 23.8 points.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Minshuo Chen, Philip Torr, Qi Zhu, Ruochen Jiao, Simon Sinong Zhan, Sipeng Chen, Yijiang Li, Zhaoran Wang, Zhenfei Yin, Zhenyang Ni","submitted_at":"2026-05-14T02:18:58Z","abstract_excerpt":"Video generation models trained on heterogeneous data with likelihood-surrogate objectives can produce visually plausible rollouts that violate physical constraints in embodied manipulation. Although reinforcement-learning post-training offers a natural route to adapting VGMs, existing video-RL rewards often reduce each rollout to a low-level visual metric, whereas manipulation video evaluation requires logic-based verification of whether the rollout satisfies a compositional task specification. To fill this gap, we introduce a compositional constraint-based reward model for post-training embo"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"CreFlow yields reward judgments better aligned with human and simulator success labels than existing methods and improves downstream execution success by 23.8 percentage points across eight bimanual manipulation tasks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the automatically formulated LTL constraints provide faithful, localized rewards without significant manual engineering or domain-specific tuning, and that the corrective reflow loss reliably stabilizes high-dimensional video diffusion updates in practice.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CreFlow combines LTL compositional rewards with credit-aware NFT and corrective reflow losses in online RL to improve embodied video diffusion models, raising downstream task success by 23.8 percentage points on eight bimanual manipulation tasks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"CreFlow uses automatically generated Linear Temporal Logic rewards plus corrective reflow to align video diffusion rollouts with embodied task rules and lift downstream success 23.8 points.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"5405d3fb8d76c379c503dbd93c2b52aef0705e2d5dd43454a5346bafcef8f514"},"source":{"id":"2605.14274","kind":"arxiv","version":1},"verdict":{"id":"aef9cafb-2caf-4f66-bc8b-d1974185da61","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T02:39:41.086820Z","strongest_claim":"CreFlow yields reward judgments better aligned with human and simulator success labels than existing methods and improves downstream execution success by 23.8 percentage points across eight bimanual manipulation tasks.","one_line_summary":"CreFlow combines LTL compositional rewards with credit-aware NFT and corrective reflow losses in online RL to improve embodied video diffusion models, raising downstream task success by 23.8 percentage points on eight bimanual manipulation tasks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the automatically formulated LTL constraints provide faithful, localized rewards without significant manual engineering or domain-specific tuning, and that the corrective reflow loss reliably stabilizes high-dimensional video diffusion updates in practice.","pith_extraction_headline":"CreFlow uses automatically generated Linear Temporal Logic rewards plus corrective reflow to align video diffusion rollouts with embodied task rules and lift downstream success 23.8 points."},"references":{"count":45,"sample":[{"doi":"","year":2025,"title":"Cosmos World Foundation Model Platform for Physical AI","work_id":"a2dba24c-318d-476a-8b21-4289c265810c","ref_index":1,"cited_arxiv_id":"2501.03575","is_internal_anchor":true},{"doi":"","year":2024,"title":"Gen2Act: Human Video Generation in Novel Scenarios enables Generalizable Robot Manipulation","work_id":"a3bde288-aace-40db-8067-3ae6656f9509","ref_index":2,"cited_arxiv_id":"2409.16283","is_internal_anchor":true},{"doi":"","year":2025,"title":"Motus: A Unified Latent Action World Model","work_id":"d0b2d257-524d-4d67-9daf-5fb43e5e977a","ref_index":3,"cited_arxiv_id":"2512.13030","is_internal_anchor":true},{"doi":"","year":2025,"title":"$\\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization","work_id":"d1ad7304-d09a-49bc-809e-846439f6aff9","ref_index":4,"cited_arxiv_id":"2504.16054","is_internal_anchor":true},{"doi":"","year":2025,"title":"SAM 3: Segment Anything with Concepts","work_id":"4a72a006-2592-4554-aad0-a9c41a9f952d","ref_index":5,"cited_arxiv_id":"2511.16719","is_internal_anchor":true}],"resolved_work":45,"snapshot_sha256":"be34ede198d7cb0d7c3623048847cb2d8608395687bd45f9d6bafc3afd597210","internal_anchors":22},"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"},"verdict_id":"aef9cafb-2caf-4f66-bc8b-d1974185da61"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DO+CODXvjP3CxTSUNLIwS67qVH7u357ljEDeuZJumr2Lij0lGvqu8itn7eKj0v2nDhcGtWljXPXSmbbsA84HBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T20:22:43.973374Z"},"content_sha256":"3566c07222a50df8826bf2944ff447043fc5d1c3e0dc9cd3e8b3281cb767ce43","schema_version":"1.0","event_id":"sha256:3566c07222a50df8826bf2944ff447043fc5d1c3e0dc9cd3e8b3281cb767ce43"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6XAIUR2GYONFPTZCI4R7XQDBSB/bundle.json","state_url":"https://pith.science/pith/6XAIUR2GYONFPTZCI4R7XQDBSB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6XAIUR2GYONFPTZCI4R7XQDBSB/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-25T20:22:43Z","links":{"resolver":"https://pith.science/pith/6XAIUR2GYONFPTZCI4R7XQDBSB","bundle":"https://pith.science/pith/6XAIUR2GYONFPTZCI4R7XQDBSB/bundle.json","state":"https://pith.science/pith/6XAIUR2GYONFPTZCI4R7XQDBSB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6XAIUR2GYONFPTZCI4R7XQDBSB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:6XAIUR2GYONFPTZCI4R7XQDBSB","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"37ec7e054359b823a87cddfe110a26171fcb72f316bcafd8535a8a9165226c43","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T02:18:58Z","title_canon_sha256":"1b215cebce7d0e38ab236f9d44487c275a2e490f5694c52f84d4a62268eba5b7"},"schema_version":"1.0","source":{"id":"2605.14274","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14274","created_at":"2026-05-17T23:39:10Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14274v1","created_at":"2026-05-17T23:39:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14274","created_at":"2026-05-17T23:39:10Z"},{"alias_kind":"pith_short_12","alias_value":"6XAIUR2GYONF","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"6XAIUR2GYONFPTZC","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"6XAIUR2G","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:3566c07222a50df8826bf2944ff447043fc5d1c3e0dc9cd3e8b3281cb767ce43","target":"graph","created_at":"2026-05-17T23:39:10Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"CreFlow yields reward judgments better aligned with human and simulator success labels than existing methods and improves downstream execution success by 23.8 percentage points across eight bimanual manipulation tasks."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the automatically formulated LTL constraints provide faithful, localized rewards without significant manual engineering or domain-specific tuning, and that the corrective reflow loss reliably stabilizes high-dimensional video diffusion updates in practice."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"CreFlow combines LTL compositional rewards with credit-aware NFT and corrective reflow losses in online RL to improve embodied video diffusion models, raising downstream task success by 23.8 percentage points on eight bimanual manipulation tasks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"CreFlow uses automatically generated Linear Temporal Logic rewards plus corrective reflow to align video diffusion rollouts with embodied task rules and lift downstream success 23.8 points."}],"snapshot_sha256":"5405d3fb8d76c379c503dbd93c2b52aef0705e2d5dd43454a5346bafcef8f514"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Video generation models trained on heterogeneous data with likelihood-surrogate objectives can produce visually plausible rollouts that violate physical constraints in embodied manipulation. Although reinforcement-learning post-training offers a natural route to adapting VGMs, existing video-RL rewards often reduce each rollout to a low-level visual metric, whereas manipulation video evaluation requires logic-based verification of whether the rollout satisfies a compositional task specification. To fill this gap, we introduce a compositional constraint-based reward model for post-training embo","authors_text":"Minshuo Chen, Philip Torr, Qi Zhu, Ruochen Jiao, Simon Sinong Zhan, Sipeng Chen, Yijiang Li, Zhaoran Wang, Zhenfei Yin, Zhenyang Ni","cross_cats":[],"headline":"CreFlow uses automatically generated Linear Temporal Logic rewards plus corrective reflow to align video diffusion rollouts with embodied task rules and lift downstream success 23.8 points.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T02:18:58Z","title":"CreFlow: Corrective Reflow for Sparse-Reward Embodied Video Diffusion RL"},"references":{"count":45,"internal_anchors":22,"resolved_work":45,"sample":[{"cited_arxiv_id":"2501.03575","doi":"","is_internal_anchor":true,"ref_index":1,"title":"Cosmos World Foundation Model Platform for Physical AI","work_id":"a2dba24c-318d-476a-8b21-4289c265810c","year":2025},{"cited_arxiv_id":"2409.16283","doi":"","is_internal_anchor":true,"ref_index":2,"title":"Gen2Act: Human Video Generation in Novel Scenarios enables Generalizable Robot Manipulation","work_id":"a3bde288-aace-40db-8067-3ae6656f9509","year":2024},{"cited_arxiv_id":"2512.13030","doi":"","is_internal_anchor":true,"ref_index":3,"title":"Motus: A Unified Latent Action World Model","work_id":"d0b2d257-524d-4d67-9daf-5fb43e5e977a","year":2025},{"cited_arxiv_id":"2504.16054","doi":"","is_internal_anchor":true,"ref_index":4,"title":"$\\pi_{0.5}$: a Vision-Language-Action Model with Open-World Generalization","work_id":"d1ad7304-d09a-49bc-809e-846439f6aff9","year":2025},{"cited_arxiv_id":"2511.16719","doi":"","is_internal_anchor":true,"ref_index":5,"title":"SAM 3: Segment Anything with Concepts","work_id":"4a72a006-2592-4554-aad0-a9c41a9f952d","year":2025}],"snapshot_sha256":"be34ede198d7cb0d7c3623048847cb2d8608395687bd45f9d6bafc3afd597210"},"source":{"id":"2605.14274","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T02:39:41.086820Z","id":"aef9cafb-2caf-4f66-bc8b-d1974185da61","model_set":{"reader":"grok-4.3"},"one_line_summary":"CreFlow combines LTL compositional rewards with credit-aware NFT and corrective reflow losses in online RL to improve embodied video diffusion models, raising downstream task success by 23.8 percentage points on eight bimanual manipulation tasks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"CreFlow uses automatically generated Linear Temporal Logic rewards plus corrective reflow to align video diffusion rollouts with embodied task rules and lift downstream success 23.8 points.","strongest_claim":"CreFlow yields reward judgments better aligned with human and simulator success labels than existing methods and improves downstream execution success by 23.8 percentage points across eight bimanual manipulation tasks.","weakest_assumption":"That the automatically formulated LTL constraints provide faithful, localized rewards without significant manual engineering or domain-specific tuning, and that the corrective reflow loss reliably stabilizes high-dimensional video diffusion updates in practice."}},"verdict_id":"aef9cafb-2caf-4f66-bc8b-d1974185da61"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:bf1f782dcfde35992974afaa166c3e60b36df13cb570e273b4f9d89a4bd3ae63","target":"record","created_at":"2026-05-17T23:39:10Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"37ec7e054359b823a87cddfe110a26171fcb72f316bcafd8535a8a9165226c43","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T02:18:58Z","title_canon_sha256":"1b215cebce7d0e38ab236f9d44487c275a2e490f5694c52f84d4a62268eba5b7"},"schema_version":"1.0","source":{"id":"2605.14274","kind":"arxiv","version":1}},"canonical_sha256":"f5c08a4746c39a57cf224723fbc061904754c270534347d31de2c1010049c53d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f5c08a4746c39a57cf224723fbc061904754c270534347d31de2c1010049c53d","first_computed_at":"2026-05-17T23:39:10.362385Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:10.362385Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1uOy3fas62FfRGgehF44zZbuFBaRK8bMArkAjSGH25mxFaxpYBzT3jJ32crAZ6qpWm0hcuG4apieGcRN+YZ1Aw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:10.362867Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14274","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bf1f782dcfde35992974afaa166c3e60b36df13cb570e273b4f9d89a4bd3ae63","sha256:3566c07222a50df8826bf2944ff447043fc5d1c3e0dc9cd3e8b3281cb767ce43"],"state_sha256":"02a867833da6606b1d78b7d2562213f39118c77e3bbc914a17af7d383904e232"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+0hjgNyN65X8lpLtivtr4pn6WGaVLRjZTiK5+GLjRF7z1idh66vOeUvCTW+qFkaqXW7ZCEhnExKgsiefXYhyCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T20:22:43.978232Z","bundle_sha256":"7fc49cc80e78b2e57fbd51456d370ce62d2add3df82d5484a2c8e8b85b6c0e02"}}