{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MIQKA47NUSNN5NOTYJTEMBF5TC","short_pith_number":"pith:MIQKA47N","schema_version":"1.0","canonical_sha256":"6220a073eda49adeb5d3c2664604bd9888e36ffb535800d01cdf92d9287d621d","source":{"kind":"arxiv","id":"2606.24884","version":1},"attestation_state":"computed","paper":{"title":"InSight: Self-Guided Skill Acquisition via Steerable VLAs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Jiajun Wu, Lars Osterberg, Mac Schwager, Maggie Wang, Ola Shorinwa, Stephen Tian","submitted_at":"2026-06-23T17:59:01Z","abstract_excerpt":"Vision-language-action (VLA) models can learn manipulation skills from demonstrations, but their capabilities are bounded by the skills in the training data. We present InSight, a framework that unlocks autonomous skill acquisition by rendering VLAs steerable at the primitive-action level (e.g., \"move gripper to the bowl\", \"lift upward\", \"pour the bottle\"). InSight consists of two primary stages: (1) an automated segmentation pipeline that partitions demonstrations into labeled primitives via VLM plan decomposition and end-effector poses to enable VLA primitive steerability, and (2) a VLM-guid"},"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":"2606.24884","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-06-23T17:59:01Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"72b4a8010bc51861a706b6ef65f4eafbabd3e812ac0302a4d94e1500d62ddc6a","abstract_canon_sha256":"1743797c0bf0b815763c9b8f475d8858212337a62552d97ec07be11337b23a22"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-24T01:15:45.383533Z","signature_b64":"vdPShil2F1tEAaxBEHaf98D0AuX+BHdau8WHZ2t/4rb+F7SM2t58LcGRRd3tZkseWtsWVkgX8rx5qcXpq2cCBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6220a073eda49adeb5d3c2664604bd9888e36ffb535800d01cdf92d9287d621d","last_reissued_at":"2026-06-24T01:15:45.383137Z","signature_status":"signed_v1","first_computed_at":"2026-06-24T01:15:45.383137Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"InSight: Self-Guided Skill Acquisition via Steerable VLAs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Jiajun Wu, Lars Osterberg, Mac Schwager, Maggie Wang, Ola Shorinwa, Stephen Tian","submitted_at":"2026-06-23T17:59:01Z","abstract_excerpt":"Vision-language-action (VLA) models can learn manipulation skills from demonstrations, but their capabilities are bounded by the skills in the training data. We present InSight, a framework that unlocks autonomous skill acquisition by rendering VLAs steerable at the primitive-action level (e.g., \"move gripper to the bowl\", \"lift upward\", \"pour the bottle\"). InSight consists of two primary stages: (1) an automated segmentation pipeline that partitions demonstrations into labeled primitives via VLM plan decomposition and end-effector poses to enable VLA primitive steerability, and (2) a VLM-guid"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.24884","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/2606.24884/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":"2606.24884","created_at":"2026-06-24T01:15:45.383192+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.24884v1","created_at":"2026-06-24T01:15:45.383192+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.24884","created_at":"2026-06-24T01:15:45.383192+00:00"},{"alias_kind":"pith_short_12","alias_value":"MIQKA47NUSNN","created_at":"2026-06-24T01:15:45.383192+00:00"},{"alias_kind":"pith_short_16","alias_value":"MIQKA47NUSNN5NOT","created_at":"2026-06-24T01:15:45.383192+00:00"},{"alias_kind":"pith_short_8","alias_value":"MIQKA47N","created_at":"2026-06-24T01:15:45.383192+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/MIQKA47NUSNN5NOTYJTEMBF5TC","json":"https://pith.science/pith/MIQKA47NUSNN5NOTYJTEMBF5TC.json","graph_json":"https://pith.science/api/pith-number/MIQKA47NUSNN5NOTYJTEMBF5TC/graph.json","events_json":"https://pith.science/api/pith-number/MIQKA47NUSNN5NOTYJTEMBF5TC/events.json","paper":"https://pith.science/paper/MIQKA47N"},"agent_actions":{"view_html":"https://pith.science/pith/MIQKA47NUSNN5NOTYJTEMBF5TC","download_json":"https://pith.science/pith/MIQKA47NUSNN5NOTYJTEMBF5TC.json","view_paper":"https://pith.science/paper/MIQKA47N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.24884&json=true","fetch_graph":"https://pith.science/api/pith-number/MIQKA47NUSNN5NOTYJTEMBF5TC/graph.json","fetch_events":"https://pith.science/api/pith-number/MIQKA47NUSNN5NOTYJTEMBF5TC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MIQKA47NUSNN5NOTYJTEMBF5TC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MIQKA47NUSNN5NOTYJTEMBF5TC/action/storage_attestation","attest_author":"https://pith.science/pith/MIQKA47NUSNN5NOTYJTEMBF5TC/action/author_attestation","sign_citation":"https://pith.science/pith/MIQKA47NUSNN5NOTYJTEMBF5TC/action/citation_signature","submit_replication":"https://pith.science/pith/MIQKA47NUSNN5NOTYJTEMBF5TC/action/replication_record"}},"created_at":"2026-06-24T01:15:45.383192+00:00","updated_at":"2026-06-24T01:15:45.383192+00:00"}