{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2AEHXCOWLCKCZRMEUDJASB3C2V","short_pith_number":"pith:2AEHXCOW","schema_version":"1.0","canonical_sha256":"d0087b89d658942cc584a0d2090762d56f2fb68b2993ad119962dd655da82bee","source":{"kind":"arxiv","id":"2503.00779","version":2},"attestation_state":"computed","paper":{"title":"Phantom: Training Robots Without Robots Using Only Human Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Jeannette Bohg, Jiaying Fang, Marion Lepert","submitted_at":"2025-03-02T08:06:55Z","abstract_excerpt":"Training general-purpose robots requires learning from large and diverse data sources. Current approaches rely heavily on teleoperated demonstrations which are difficult to scale. We present a scalable framework for training manipulation policies directly from human video demonstrations, requiring no robot data. Our method converts human demonstrations into robot-compatible observation-action pairs using hand pose estimation and visual data editing. We inpaint the human arm and overlay a rendered robot to align the visual domains. This enables zero-shot deployment on real hardware without any "},"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":"2503.00779","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-03-02T08:06:55Z","cross_cats_sorted":[],"title_canon_sha256":"2e5c107c9c29623d09b489f584774c0c8461c36832fa1fb30d3ed677b5c214c7","abstract_canon_sha256":"349a8c6671c162c7245d5ca7017b2a34580448ae9b98d529c219e4edbbcf8ff4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:04:52.263687Z","signature_b64":"OBMmW0OElwasKiHt/7IWr03hSiebeYYCksDLAvB7psGDvXeci7zf03xJPeG4FX3Ees750H1T14gyuFdQJDISDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d0087b89d658942cc584a0d2090762d56f2fb68b2993ad119962dd655da82bee","last_reissued_at":"2026-05-29T01:04:52.263233Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:04:52.263233Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Phantom: Training Robots Without Robots Using Only Human Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Jeannette Bohg, Jiaying Fang, Marion Lepert","submitted_at":"2025-03-02T08:06:55Z","abstract_excerpt":"Training general-purpose robots requires learning from large and diverse data sources. Current approaches rely heavily on teleoperated demonstrations which are difficult to scale. We present a scalable framework for training manipulation policies directly from human video demonstrations, requiring no robot data. Our method converts human demonstrations into robot-compatible observation-action pairs using hand pose estimation and visual data editing. We inpaint the human arm and overlay a rendered robot to align the visual domains. This enables zero-shot deployment on real hardware without any "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.00779","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/2503.00779/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":"2503.00779","created_at":"2026-05-29T01:04:52.263300+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.00779v2","created_at":"2026-05-29T01:04:52.263300+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.00779","created_at":"2026-05-29T01:04:52.263300+00:00"},{"alias_kind":"pith_short_12","alias_value":"2AEHXCOWLCKC","created_at":"2026-05-29T01:04:52.263300+00:00"},{"alias_kind":"pith_short_16","alias_value":"2AEHXCOWLCKCZRME","created_at":"2026-05-29T01:04:52.263300+00:00"},{"alias_kind":"pith_short_8","alias_value":"2AEHXCOW","created_at":"2026-05-29T01:04:52.263300+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":12,"sample":[{"citing_arxiv_id":"2605.16743","citing_title":"LACE: Latent Visual Representation for Cross-Embodiment Learning","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2507.00990","citing_title":"Robotic Manipulation by Imitating Generated Videos Without Physical Demonstrations","ref_index":65,"is_internal_anchor":true},{"citing_arxiv_id":"2510.08547","citing_title":"R2RGEN: Real-to-Real 3D Data Generation for Spatially Generalized Manipulation","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2511.04671","citing_title":"X-Diffusion: Training Diffusion Policies on Cross-Embodiment Human Demonstrations","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2511.17001","citing_title":"Unify Robot Actions in Camera Frame","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2512.01773","citing_title":"IGen: Scalable Data Generation for Robot Learning from Open-World Images","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19092","citing_title":"RoboWM-Bench: A Benchmark for Evaluating World Models in Robotic Manipulation","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12038","citing_title":"OmniHumanoid: Streaming Cross-Embodiment Video Generation with Paired-Free Adaptation","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2605.03637","citing_title":"Bridging the Embodiment Gap: Disentangled Cross-Embodiment Video Editing","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2604.10809","citing_title":"WARPED: Wrist-Aligned Rendering for Robot Policy Learning from Egocentric Human Demonstrations","ref_index":60,"is_internal_anchor":true},{"citing_arxiv_id":"2604.07607","citing_title":"EgoVerse: An Egocentric Human Dataset for Robot Learning from Around the World","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19092","citing_title":"RoboWM-Bench: A Benchmark for Evaluating World Models in Robotic Manipulation","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2AEHXCOWLCKCZRMEUDJASB3C2V","json":"https://pith.science/pith/2AEHXCOWLCKCZRMEUDJASB3C2V.json","graph_json":"https://pith.science/api/pith-number/2AEHXCOWLCKCZRMEUDJASB3C2V/graph.json","events_json":"https://pith.science/api/pith-number/2AEHXCOWLCKCZRMEUDJASB3C2V/events.json","paper":"https://pith.science/paper/2AEHXCOW"},"agent_actions":{"view_html":"https://pith.science/pith/2AEHXCOWLCKCZRMEUDJASB3C2V","download_json":"https://pith.science/pith/2AEHXCOWLCKCZRMEUDJASB3C2V.json","view_paper":"https://pith.science/paper/2AEHXCOW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.00779&json=true","fetch_graph":"https://pith.science/api/pith-number/2AEHXCOWLCKCZRMEUDJASB3C2V/graph.json","fetch_events":"https://pith.science/api/pith-number/2AEHXCOWLCKCZRMEUDJASB3C2V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2AEHXCOWLCKCZRMEUDJASB3C2V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2AEHXCOWLCKCZRMEUDJASB3C2V/action/storage_attestation","attest_author":"https://pith.science/pith/2AEHXCOWLCKCZRMEUDJASB3C2V/action/author_attestation","sign_citation":"https://pith.science/pith/2AEHXCOWLCKCZRMEUDJASB3C2V/action/citation_signature","submit_replication":"https://pith.science/pith/2AEHXCOWLCKCZRMEUDJASB3C2V/action/replication_record"}},"created_at":"2026-05-29T01:04:52.263300+00:00","updated_at":"2026-05-29T01:04:52.263300+00:00"}