{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:EIZWMWEUMZCMUBZLTG67IBBKQQ","short_pith_number":"pith:EIZWMWEU","schema_version":"1.0","canonical_sha256":"22336658946644ca072b99bdf4042a84237c7789b4d1a7cd39e687024dbc2bfe","source":{"kind":"arxiv","id":"2605.17077","version":1},"attestation_state":"computed","paper":{"title":"How to Instruct Your Robot: Dense Language Annotations Power Robot Policy Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Alexander Trevithick, Bosung Kim, Brandon Cui, David Acuna, Jaehun Jung, Prithviraj Ammanabrolu, Ruiyi Wang, Yejin Choi","submitted_at":"2026-05-16T16:52:08Z","abstract_excerpt":"Scaling robot policy learning is bottlenecked by the cost of collecting demonstrations, while language annotations for existing demonstrations are comparatively cheap. We study language density as a lever for extracting more signal from a fixed robot or egocentric-video corpus. We introduce DeMiAn (Dense Multi-aspect Annotation), a two-stage approach that first re-labels demonstration segments with VLM-generated annotations along four complementary aspects: physical motion, scene composition, arm pose, and reasoning. A learned instructor then maps a task description and initial scene snapshot "},"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":"2605.17077","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-05-16T16:52:08Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4864d69f607cdef0e008a1943c255837aefd9a055f7e9ba78719c4ab1f1620d6","abstract_canon_sha256":"3ff857c05f8145cb6bd034473a5c45adf8e6cfb7e14898040dd20086ee0e8e8b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:03:39.384387Z","signature_b64":"YQmd1YawmWxCPSOkeFM7bNqGiYUH+ucgY4adIoaoeO/yIeaqnMRTpdIWqBb3O+PgHQCMqx7y5VzV5W/JJvR+DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"22336658946644ca072b99bdf4042a84237c7789b4d1a7cd39e687024dbc2bfe","last_reissued_at":"2026-05-20T00:03:39.383830Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:03:39.383830Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How to Instruct Your Robot: Dense Language Annotations Power Robot Policy Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Alexander Trevithick, Bosung Kim, Brandon Cui, David Acuna, Jaehun Jung, Prithviraj Ammanabrolu, Ruiyi Wang, Yejin Choi","submitted_at":"2026-05-16T16:52:08Z","abstract_excerpt":"Scaling robot policy learning is bottlenecked by the cost of collecting demonstrations, while language annotations for existing demonstrations are comparatively cheap. We study language density as a lever for extracting more signal from a fixed robot or egocentric-video corpus. We introduce DeMiAn (Dense Multi-aspect Annotation), a two-stage approach that first re-labels demonstration segments with VLM-generated annotations along four complementary aspects: physical motion, scene composition, arm pose, and reasoning. A learned instructor then maps a task description and initial scene snapshot "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17077","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/2605.17077/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-19T22:33:23.810245Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T22:21:57.747581Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"ab642f6e85a451e8c84ad2eefc9179d8813be331e533d1b64d23a07d9306b3aa"},"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":"2605.17077","created_at":"2026-05-20T00:03:39.383915+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.17077v1","created_at":"2026-05-20T00:03:39.383915+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17077","created_at":"2026-05-20T00:03:39.383915+00:00"},{"alias_kind":"pith_short_12","alias_value":"EIZWMWEUMZCM","created_at":"2026-05-20T00:03:39.383915+00:00"},{"alias_kind":"pith_short_16","alias_value":"EIZWMWEUMZCMUBZL","created_at":"2026-05-20T00:03:39.383915+00:00"},{"alias_kind":"pith_short_8","alias_value":"EIZWMWEU","created_at":"2026-05-20T00:03:39.383915+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/EIZWMWEUMZCMUBZLTG67IBBKQQ","json":"https://pith.science/pith/EIZWMWEUMZCMUBZLTG67IBBKQQ.json","graph_json":"https://pith.science/api/pith-number/EIZWMWEUMZCMUBZLTG67IBBKQQ/graph.json","events_json":"https://pith.science/api/pith-number/EIZWMWEUMZCMUBZLTG67IBBKQQ/events.json","paper":"https://pith.science/paper/EIZWMWEU"},"agent_actions":{"view_html":"https://pith.science/pith/EIZWMWEUMZCMUBZLTG67IBBKQQ","download_json":"https://pith.science/pith/EIZWMWEUMZCMUBZLTG67IBBKQQ.json","view_paper":"https://pith.science/paper/EIZWMWEU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.17077&json=true","fetch_graph":"https://pith.science/api/pith-number/EIZWMWEUMZCMUBZLTG67IBBKQQ/graph.json","fetch_events":"https://pith.science/api/pith-number/EIZWMWEUMZCMUBZLTG67IBBKQQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EIZWMWEUMZCMUBZLTG67IBBKQQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EIZWMWEUMZCMUBZLTG67IBBKQQ/action/storage_attestation","attest_author":"https://pith.science/pith/EIZWMWEUMZCMUBZLTG67IBBKQQ/action/author_attestation","sign_citation":"https://pith.science/pith/EIZWMWEUMZCMUBZLTG67IBBKQQ/action/citation_signature","submit_replication":"https://pith.science/pith/EIZWMWEUMZCMUBZLTG67IBBKQQ/action/replication_record"}},"created_at":"2026-05-20T00:03:39.383915+00:00","updated_at":"2026-05-20T00:03:39.383915+00:00"}