{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TS64HG6KOSTH3FIONHPEGCAHKL","short_pith_number":"pith:TS64HG6K","schema_version":"1.0","canonical_sha256":"9cbdc39bca74a67d950e69de43080752efa40b602e04122fecfc3facd72c132d","source":{"kind":"arxiv","id":"2605.16241","version":1},"attestation_state":"computed","paper":{"title":"Offline Semantic Guidance for Efficient Vision-Language-Action Policy Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Brady Zhang, Jin Shi, Yishun Lu","submitted_at":"2026-05-15T17:48:25Z","abstract_excerpt":"Billion-parameter Vision-Language-Action (VLA) policies have recently shown impressive performance in robotic manipulation, yet their size and inference cost remain major obstacles for real-time closed-loop control. We introduce \\textbf{VLA-AD}, a distillation framework that uses a Vision-Language Model as an offline semantic supervisor to transfer large VLA teachers into lightweight student policies. Instead of relying only on low-level action imitation, VLA-AD augments teacher-provided 7-DoF action targets with high-level semantic guidance, including task phase anchors and multi-frame operat"},"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":"2605.16241","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-15T17:48:25Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e4faf20917be8e16de3299bc108dc0f70af25f8a62ae2f597e1cd1420c169307","abstract_canon_sha256":"1cc27bfdbe8d5333c03d0f0bfd51a990793f18c2b5205ac4894c7c6c3ffcd67d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:01:59.659900Z","signature_b64":"Q+AQdm+UH8Qbjo1OYCqmLr2L2UBjoiUcYh6cJFBtG2CLSt4+8ztsz3stS9EajqEw40w8nECLQW9k3EqQZ+QkCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9cbdc39bca74a67d950e69de43080752efa40b602e04122fecfc3facd72c132d","last_reissued_at":"2026-05-20T00:01:59.659064Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:01:59.659064Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Offline Semantic Guidance for Efficient Vision-Language-Action Policy Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Brady Zhang, Jin Shi, Yishun Lu","submitted_at":"2026-05-15T17:48:25Z","abstract_excerpt":"Billion-parameter Vision-Language-Action (VLA) policies have recently shown impressive performance in robotic manipulation, yet their size and inference cost remain major obstacles for real-time closed-loop control. We introduce \\textbf{VLA-AD}, a distillation framework that uses a Vision-Language Model as an offline semantic supervisor to transfer large VLA teachers into lightweight student policies. Instead of relying only on low-level action imitation, VLA-AD augments teacher-provided 7-DoF action targets with high-level semantic guidance, including task phase anchors and multi-frame operat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.16241","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/2605.16241/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"shingle_duplication","ran_at":"2026-05-19T17:49:42.190933Z","status":"skipped","version":"0.1.0","findings_count":0},{"name":"citation_quote_validity","ran_at":"2026-05-19T17:49:41.799961Z","status":"skipped","version":"0.1.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T17:33:23.105605Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"external_links","ran_at":"2026-05-19T17:31:26.215193Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T17:01:55.612193Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"cited_work_retraction","ran_at":"2026-05-19T16:51:57.401601Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"fca234e583c4957242f28136adb227436e0eeda161ca141c0bd24830b461ed1e"},"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":"2605.16241","created_at":"2026-05-20T00:01:59.659189+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.16241v1","created_at":"2026-05-20T00:01:59.659189+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16241","created_at":"2026-05-20T00:01:59.659189+00:00"},{"alias_kind":"pith_short_12","alias_value":"TS64HG6KOSTH","created_at":"2026-05-20T00:01:59.659189+00:00"},{"alias_kind":"pith_short_16","alias_value":"TS64HG6KOSTH3FIO","created_at":"2026-05-20T00:01:59.659189+00:00"},{"alias_kind":"pith_short_8","alias_value":"TS64HG6K","created_at":"2026-05-20T00:01:59.659189+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/TS64HG6KOSTH3FIONHPEGCAHKL","json":"https://pith.science/pith/TS64HG6KOSTH3FIONHPEGCAHKL.json","graph_json":"https://pith.science/api/pith-number/TS64HG6KOSTH3FIONHPEGCAHKL/graph.json","events_json":"https://pith.science/api/pith-number/TS64HG6KOSTH3FIONHPEGCAHKL/events.json","paper":"https://pith.science/paper/TS64HG6K"},"agent_actions":{"view_html":"https://pith.science/pith/TS64HG6KOSTH3FIONHPEGCAHKL","download_json":"https://pith.science/pith/TS64HG6KOSTH3FIONHPEGCAHKL.json","view_paper":"https://pith.science/paper/TS64HG6K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.16241&json=true","fetch_graph":"https://pith.science/api/pith-number/TS64HG6KOSTH3FIONHPEGCAHKL/graph.json","fetch_events":"https://pith.science/api/pith-number/TS64HG6KOSTH3FIONHPEGCAHKL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TS64HG6KOSTH3FIONHPEGCAHKL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TS64HG6KOSTH3FIONHPEGCAHKL/action/storage_attestation","attest_author":"https://pith.science/pith/TS64HG6KOSTH3FIONHPEGCAHKL/action/author_attestation","sign_citation":"https://pith.science/pith/TS64HG6KOSTH3FIONHPEGCAHKL/action/citation_signature","submit_replication":"https://pith.science/pith/TS64HG6KOSTH3FIONHPEGCAHKL/action/replication_record"}},"created_at":"2026-05-20T00:01:59.659189+00:00","updated_at":"2026-05-20T00:01:59.659189+00:00"}