{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:XZRI52YHZDBK3MXK4TZBQD3IA6","short_pith_number":"pith:XZRI52YH","canonical_record":{"source":{"id":"2607.27782","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-07-30T07:14:39Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6c685f69a2651f0169ec27c2bca09db06d14b2bdb87aff83ef29a6555850e50f","abstract_canon_sha256":"1133e2491c1ef9e4118cc9f487dbd634664fca97fc6c47bb61d9cd0de17e097f"},"schema_version":"1.0"},"canonical_sha256":"be628eeb07c8c2adb2eae4f2180f6807a94e3900e0a2a4a0678c22ce47e1f7b2","source":{"kind":"arxiv","id":"2607.27782","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.27782","created_at":"2026-07-31T01:30:46Z"},{"alias_kind":"arxiv_version","alias_value":"2607.27782v1","created_at":"2026-07-31T01:30:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.27782","created_at":"2026-07-31T01:30:46Z"},{"alias_kind":"pith_short_12","alias_value":"XZRI52YHZDBK","created_at":"2026-07-31T01:30:46Z"},{"alias_kind":"pith_short_16","alias_value":"XZRI52YHZDBK3MXK","created_at":"2026-07-31T01:30:46Z"},{"alias_kind":"pith_short_8","alias_value":"XZRI52YH","created_at":"2026-07-31T01:30:46Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:XZRI52YHZDBK3MXK4TZBQD3IA6","target":"record","payload":{"canonical_record":{"source":{"id":"2607.27782","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-07-30T07:14:39Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6c685f69a2651f0169ec27c2bca09db06d14b2bdb87aff83ef29a6555850e50f","abstract_canon_sha256":"1133e2491c1ef9e4118cc9f487dbd634664fca97fc6c47bb61d9cd0de17e097f"},"schema_version":"1.0"},"canonical_sha256":"be628eeb07c8c2adb2eae4f2180f6807a94e3900e0a2a4a0678c22ce47e1f7b2","receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be628eeb07c8c2adb2eae4f2180f6807a94e3900e0a2a4a0678c22ce47e1f7b2","last_reissued_at":"2026-07-31T01:30:46.182727Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-31T01:30:46.182727Z"},"source_kind":"arxiv","source_id":"2607.27782","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":"2026-07-31T01:30:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3TU2ZswellFbzZKEiLBBVp8zUifV8SyRYFAt8Bp/FNjGA1EBi3wF7tZZdK74oiWcv2ELbOELXDtO5KUt3Le+Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:15:07.186923Z"},"content_sha256":"ad77b12ed05b05af10c89521875331a3f695c9efc32bdd846b88579744e7e4c6","schema_version":"1.0","event_id":"sha256:ad77b12ed05b05af10c89521875331a3f695c9efc32bdd846b88579744e7e4c6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:XZRI52YHZDBK3MXK4TZBQD3IA6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RedFlow: Redirect Failure into Action-Level Corrections for Flow-matching VLA Policy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Fangqi Zhu, Junhao Li, Quanxin Shou, Song Guo, Xiaoyi Pang, Yikun Miao, Zhengyang Yan, Zicong Hong, Zijun Wang","submitted_at":"2026-07-30T07:14:39Z","abstract_excerpt":"Flow-matching Vision-Language-Action (VLA) policies have shown strong potential for robotic manipulation but often suffer from compounding errors caused by distribution shifts during deployment. While offline reinforcement learning (RL) provides a practical way to improve deployed policies using rollout data, existing methods either ignore failure data or exploit it only at the trajectory level, resulting in low learning efficiency and persistent errors. We propose **RedFlow**, a fine-grained offline RL framework that redirects failure experiences into action-level corrective supervision for f"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.27782","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/2607.27782/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-31T01:30:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"X6OiiGSMmkV8cR3RW9/Rmjpe81Zv8VQk5bZ+/CQxrP5gj8Dw3m5l2onKOj0b1vC5Sd8eWy+89Ysd8VnwrnQ2Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:15:07.187419Z"},"content_sha256":"4449e3485318c979d52d243f5caa54431af4acb84849029242d0ac4fe43e534d","schema_version":"1.0","event_id":"sha256:4449e3485318c979d52d243f5caa54431af4acb84849029242d0ac4fe43e534d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XZRI52YHZDBK3MXK4TZBQD3IA6/bundle.json","state_url":"https://pith.science/pith/XZRI52YHZDBK3MXK4TZBQD3IA6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XZRI52YHZDBK3MXK4TZBQD3IA6/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":"2026-08-04T16:15:07Z","links":{"resolver":"https://pith.science/pith/XZRI52YHZDBK3MXK4TZBQD3IA6","bundle":"https://pith.science/pith/XZRI52YHZDBK3MXK4TZBQD3IA6/bundle.json","state":"https://pith.science/pith/XZRI52YHZDBK3MXK4TZBQD3IA6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XZRI52YHZDBK3MXK4TZBQD3IA6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:XZRI52YHZDBK3MXK4TZBQD3IA6","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":"1133e2491c1ef9e4118cc9f487dbd634664fca97fc6c47bb61d9cd0de17e097f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-07-30T07:14:39Z","title_canon_sha256":"6c685f69a2651f0169ec27c2bca09db06d14b2bdb87aff83ef29a6555850e50f"},"schema_version":"1.0","source":{"id":"2607.27782","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.27782","created_at":"2026-07-31T01:30:46Z"},{"alias_kind":"arxiv_version","alias_value":"2607.27782v1","created_at":"2026-07-31T01:30:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.27782","created_at":"2026-07-31T01:30:46Z"},{"alias_kind":"pith_short_12","alias_value":"XZRI52YHZDBK","created_at":"2026-07-31T01:30:46Z"},{"alias_kind":"pith_short_16","alias_value":"XZRI52YHZDBK3MXK","created_at":"2026-07-31T01:30:46Z"},{"alias_kind":"pith_short_8","alias_value":"XZRI52YH","created_at":"2026-07-31T01:30:46Z"}],"graph_snapshots":[{"event_id":"sha256:4449e3485318c979d52d243f5caa54431af4acb84849029242d0ac4fe43e534d","target":"graph","created_at":"2026-07-31T01:30:46Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2607.27782/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Flow-matching Vision-Language-Action (VLA) policies have shown strong potential for robotic manipulation but often suffer from compounding errors caused by distribution shifts during deployment. While offline reinforcement learning (RL) provides a practical way to improve deployed policies using rollout data, existing methods either ignore failure data or exploit it only at the trajectory level, resulting in low learning efficiency and persistent errors. We propose **RedFlow**, a fine-grained offline RL framework that redirects failure experiences into action-level corrective supervision for f","authors_text":"Fangqi Zhu, Junhao Li, Quanxin Shou, Song Guo, Xiaoyi Pang, Yikun Miao, Zhengyang Yan, Zicong Hong, Zijun Wang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-07-30T07:14:39Z","title":"RedFlow: Redirect Failure into Action-Level Corrections for Flow-matching VLA Policy"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.27782","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ad77b12ed05b05af10c89521875331a3f695c9efc32bdd846b88579744e7e4c6","target":"record","created_at":"2026-07-31T01:30:46Z","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":"1133e2491c1ef9e4118cc9f487dbd634664fca97fc6c47bb61d9cd0de17e097f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-07-30T07:14:39Z","title_canon_sha256":"6c685f69a2651f0169ec27c2bca09db06d14b2bdb87aff83ef29a6555850e50f"},"schema_version":"1.0","source":{"id":"2607.27782","kind":"arxiv","version":1}},"canonical_sha256":"be628eeb07c8c2adb2eae4f2180f6807a94e3900e0a2a4a0678c22ce47e1f7b2","receipt":{"builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"be628eeb07c8c2adb2eae4f2180f6807a94e3900e0a2a4a0678c22ce47e1f7b2","first_computed_at":"2026-07-31T01:30:46.182727Z","kind":"pith_receipt","last_reissued_at":"2026-07-31T01:30:46.182727Z","receipt_version":"0.3","signature_status":"unsigned_v0"},"source_id":"2607.27782","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ad77b12ed05b05af10c89521875331a3f695c9efc32bdd846b88579744e7e4c6","sha256:4449e3485318c979d52d243f5caa54431af4acb84849029242d0ac4fe43e534d"],"state_sha256":"106b25b3c7ffd6ca67608127a7fe0b180f97d87523c197a0d93f1f91176aaee9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wQjBV770WyrRC7yf/YSo5otqd5MGpSlhtq7FkKaV6ieF4yr8n5hxNoaTy/G8AlFFET36CCgzJ4p2P/SJ79NTAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T16:15:07.191059Z","bundle_sha256":"1a4f8568fc72bfcbe3064f4b63ad459c0f277c21590f62625788c8cc95bdbc57"}}