{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6QJIGCPXFJE3CLY2UJA6BVEDJ6","short_pith_number":"pith:6QJIGCPX","schema_version":"1.0","canonical_sha256":"f4128309f72a49b12f1aa241e0d4834f97a4d9e51f383d83bdea5f2f26ce3a7d","source":{"kind":"arxiv","id":"2509.09769","version":1},"attestation_state":"computed","paper":{"title":"MimicDroid: In-Context Learning for Humanoid Robot Manipulation from Human Play Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Mingyo Seo, Qi Wang, Roberto Mart\\'in-Mart\\'in, Rutav Shah, Sateesh Kumar, Shuijing Liu, Yuke Zhu, Zhenyu Jiang","submitted_at":"2025-09-11T18:02:48Z","abstract_excerpt":"We aim to enable humanoid robots to efficiently solve new manipulation tasks from a few video examples. In-context learning (ICL) is a promising framework for achieving this goal due to its test-time data efficiency and rapid adaptability. However, current ICL methods rely on labor-intensive teleoperated data for training, which restricts scalability. We propose using human play videos -- continuous, unlabeled videos of people interacting freely with their environment -- as a scalable and diverse training data source. We introduce MimicDroid, which enables humanoids to perform ICL using human "},"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":"2509.09769","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-09-11T18:02:48Z","cross_cats_sorted":[],"title_canon_sha256":"cd5548771d85ef9b856a63412c2463407bbc1bcae4b2224b1da959472387218d","abstract_canon_sha256":"46c0e3e704c0950425fb4c018022dd20d57e7cc47224e936ec3600b455f7dc20"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:09:48.902402Z","signature_b64":"05ZTssKeRJW1oqE9hubnXV9CbVIjdObw8IbvazzRTYSVuRVzLDz4zzYfsZnBZQPtgMp96eET0+kGFBWmRu07Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f4128309f72a49b12f1aa241e0d4834f97a4d9e51f383d83bdea5f2f26ce3a7d","last_reissued_at":"2026-07-05T12:09:48.901873Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:09:48.901873Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MimicDroid: In-Context Learning for Humanoid Robot Manipulation from Human Play Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Mingyo Seo, Qi Wang, Roberto Mart\\'in-Mart\\'in, Rutav Shah, Sateesh Kumar, Shuijing Liu, Yuke Zhu, Zhenyu Jiang","submitted_at":"2025-09-11T18:02:48Z","abstract_excerpt":"We aim to enable humanoid robots to efficiently solve new manipulation tasks from a few video examples. In-context learning (ICL) is a promising framework for achieving this goal due to its test-time data efficiency and rapid adaptability. However, current ICL methods rely on labor-intensive teleoperated data for training, which restricts scalability. We propose using human play videos -- continuous, unlabeled videos of people interacting freely with their environment -- as a scalable and diverse training data source. We introduce MimicDroid, which enables humanoids to perform ICL using human "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.09769","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/2509.09769/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":"2509.09769","created_at":"2026-07-05T12:09:48.901935+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.09769v1","created_at":"2026-07-05T12:09:48.901935+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.09769","created_at":"2026-07-05T12:09:48.901935+00:00"},{"alias_kind":"pith_short_12","alias_value":"6QJIGCPXFJE3","created_at":"2026-07-05T12:09:48.901935+00:00"},{"alias_kind":"pith_short_16","alias_value":"6QJIGCPXFJE3CLY2","created_at":"2026-07-05T12:09:48.901935+00:00"},{"alias_kind":"pith_short_8","alias_value":"6QJIGCPX","created_at":"2026-07-05T12:09:48.901935+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06988","citing_title":"WAM-TTT: Steering World-Action Models by Watching Human Play at Test Time","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2606.26025","citing_title":"In-Context World Modeling for Robotic Control","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26025","citing_title":"In-Context World Modeling for Robotic Control","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05952","citing_title":"Learning of Robot Safety Policies via Adversarial Synthetic Scenarios","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24934","citing_title":"HumanEgo: Zero-Shot Robot Learning from Minutes of Human Egocentric Videos","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2511.04671","citing_title":"X-Diffusion: Training Diffusion Policies on Cross-Embodiment Human Demonstrations","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20348","citing_title":"Bimanual Robot Manipulation via Multi-Agent In-Context Learning","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6QJIGCPXFJE3CLY2UJA6BVEDJ6","json":"https://pith.science/pith/6QJIGCPXFJE3CLY2UJA6BVEDJ6.json","graph_json":"https://pith.science/api/pith-number/6QJIGCPXFJE3CLY2UJA6BVEDJ6/graph.json","events_json":"https://pith.science/api/pith-number/6QJIGCPXFJE3CLY2UJA6BVEDJ6/events.json","paper":"https://pith.science/paper/6QJIGCPX"},"agent_actions":{"view_html":"https://pith.science/pith/6QJIGCPXFJE3CLY2UJA6BVEDJ6","download_json":"https://pith.science/pith/6QJIGCPXFJE3CLY2UJA6BVEDJ6.json","view_paper":"https://pith.science/paper/6QJIGCPX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.09769&json=true","fetch_graph":"https://pith.science/api/pith-number/6QJIGCPXFJE3CLY2UJA6BVEDJ6/graph.json","fetch_events":"https://pith.science/api/pith-number/6QJIGCPXFJE3CLY2UJA6BVEDJ6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6QJIGCPXFJE3CLY2UJA6BVEDJ6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6QJIGCPXFJE3CLY2UJA6BVEDJ6/action/storage_attestation","attest_author":"https://pith.science/pith/6QJIGCPXFJE3CLY2UJA6BVEDJ6/action/author_attestation","sign_citation":"https://pith.science/pith/6QJIGCPXFJE3CLY2UJA6BVEDJ6/action/citation_signature","submit_replication":"https://pith.science/pith/6QJIGCPXFJE3CLY2UJA6BVEDJ6/action/replication_record"}},"created_at":"2026-07-05T12:09:48.901935+00:00","updated_at":"2026-07-05T12:09:48.901935+00:00"}