{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:EHOEEKIAM4THYBFDFDDFNWC2GF","short_pith_number":"pith:EHOEEKIA","schema_version":"1.0","canonical_sha256":"21dc42290067267c04a328c656d85a317640228a51cab332882694d5e085aa30","source":{"kind":"arxiv","id":"2602.12978","version":2},"attestation_state":"computed","paper":{"title":"Learning Native Continuation for Action Chunking Flow Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Bocheng Li, Dequan Wang, Di Zhang, Hang Yu, Junliang Guo, Juntu Zhao, Junyuan Xie, Mingzhu Li, Wenxuan Wu, Yang Gao, Yingdong Hu, Yufeng Liu","submitted_at":"2026-02-13T14:56:06Z","abstract_excerpt":"Action chunking enables Vision Language Action (VLA) models to run in real time, but naive chunked execution often exhibits discontinuities at chunk boundaries. Real-Time Chunking (RTC) alleviates this issue but is external to the policy, leading to spurious multimodal switching and trajectories that are not intrinsically smooth. We propose Legato, a training-time continuation method for action-chunked flow-based VLA policies. Specifically, Legato initializes denoising from a schedule-shaped mixture of known actions and noise, exposing the model to partial action information. Moreover, Legato "},"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":"2602.12978","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-02-13T14:56:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b0d82651be0c76a2bb6ef9f34503ed5a877052bc4a48d98d6b68f24ddb0b058f","abstract_canon_sha256":"d2193789a1e9def507cdbea3e1b166249ed8ef9f2da2afa2181323d98e60395d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:25.669759Z","signature_b64":"BYmqZdmUBsJGzTCp4TVKmcJ/fID0fS3Y4FLit6BOf/PScIXR9ZqYUY/ADoDbTN7UGTNVEnhr/+soWiXbAt/BBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"21dc42290067267c04a328c656d85a317640228a51cab332882694d5e085aa30","last_reissued_at":"2026-05-20T00:04:25.668763Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:25.668763Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Native Continuation for Action Chunking Flow Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Bocheng Li, Dequan Wang, Di Zhang, Hang Yu, Junliang Guo, Juntu Zhao, Junyuan Xie, Mingzhu Li, Wenxuan Wu, Yang Gao, Yingdong Hu, Yufeng Liu","submitted_at":"2026-02-13T14:56:06Z","abstract_excerpt":"Action chunking enables Vision Language Action (VLA) models to run in real time, but naive chunked execution often exhibits discontinuities at chunk boundaries. Real-Time Chunking (RTC) alleviates this issue but is external to the policy, leading to spurious multimodal switching and trajectories that are not intrinsically smooth. We propose Legato, a training-time continuation method for action-chunked flow-based VLA policies. Specifically, Legato initializes denoising from a schedule-shaped mixture of known actions and noise, exposing the model to partial action information. Moreover, Legato "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.12978","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/2602.12978/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":"2602.12978","created_at":"2026-05-20T00:04:25.668923+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.12978v2","created_at":"2026-05-20T00:04:25.668923+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.12978","created_at":"2026-05-20T00:04:25.668923+00:00"},{"alias_kind":"pith_short_12","alias_value":"EHOEEKIAM4TH","created_at":"2026-05-20T00:04:25.668923+00:00"},{"alias_kind":"pith_short_16","alias_value":"EHOEEKIAM4THYBFD","created_at":"2026-05-20T00:04:25.668923+00:00"},{"alias_kind":"pith_short_8","alias_value":"EHOEEKIA","created_at":"2026-05-20T00:04:25.668923+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":6,"sample":[{"citing_arxiv_id":"2603.11642","citing_title":"Noise-Space Attribution and Control of Chunk-Boundary Artifact","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2603.19199","citing_title":"FASTER: Rethinking Real-Time Flow VLAs","ref_index":53,"is_internal_anchor":true},{"citing_arxiv_id":"2603.19199","citing_title":"FASTER: Rethinking Real-Time Flow VLAs","ref_index":53,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11459","citing_title":"Overcoming Dynamics-Blindness: Training-Free Pace-and-Path Correction for VLA Models","ref_index":53,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11459","citing_title":"Overcoming Dynamics-Blindness: Training-Free Pace-and-Path Correction for VLA Models","ref_index":53,"is_internal_anchor":true},{"citing_arxiv_id":"2604.25050","citing_title":"DiscreteRTC: Discrete Diffusion Policies are Natural Asynchronous Executors","ref_index":23,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EHOEEKIAM4THYBFDFDDFNWC2GF","json":"https://pith.science/pith/EHOEEKIAM4THYBFDFDDFNWC2GF.json","graph_json":"https://pith.science/api/pith-number/EHOEEKIAM4THYBFDFDDFNWC2GF/graph.json","events_json":"https://pith.science/api/pith-number/EHOEEKIAM4THYBFDFDDFNWC2GF/events.json","paper":"https://pith.science/paper/EHOEEKIA"},"agent_actions":{"view_html":"https://pith.science/pith/EHOEEKIAM4THYBFDFDDFNWC2GF","download_json":"https://pith.science/pith/EHOEEKIAM4THYBFDFDDFNWC2GF.json","view_paper":"https://pith.science/paper/EHOEEKIA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.12978&json=true","fetch_graph":"https://pith.science/api/pith-number/EHOEEKIAM4THYBFDFDDFNWC2GF/graph.json","fetch_events":"https://pith.science/api/pith-number/EHOEEKIAM4THYBFDFDDFNWC2GF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EHOEEKIAM4THYBFDFDDFNWC2GF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EHOEEKIAM4THYBFDFDDFNWC2GF/action/storage_attestation","attest_author":"https://pith.science/pith/EHOEEKIAM4THYBFDFDDFNWC2GF/action/author_attestation","sign_citation":"https://pith.science/pith/EHOEEKIAM4THYBFDFDDFNWC2GF/action/citation_signature","submit_replication":"https://pith.science/pith/EHOEEKIAM4THYBFDFDDFNWC2GF/action/replication_record"}},"created_at":"2026-05-20T00:04:25.668923+00:00","updated_at":"2026-05-20T00:04:25.668923+00:00"}