{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7QOFHK7VUBLOR6WDVGTPRLPERI","short_pith_number":"pith:7QOFHK7V","schema_version":"1.0","canonical_sha256":"fc1c53abf5a056e8fac3a9a6f8ade48a0e1d5f210a20735fce28b87760ccb49c","source":{"kind":"arxiv","id":"2407.15208","version":2},"attestation_state":"computed","paper":{"title":"Flow as the Cross-Domain Manipulation Interface","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Cheng Chi, Gordon Wetzstein, Manuela Veloso, Mengda Xu, Shuran Song, Yinghao Xu, Zhenjia Xu","submitted_at":"2024-07-21T16:15:02Z","abstract_excerpt":"We present Im2Flow2Act, a scalable learning framework that enables robots to acquire real-world manipulation skills without the need of real-world robot training data. The key idea behind Im2Flow2Act is to use object flow as the manipulation interface, bridging domain gaps between different embodiments (i.e., human and robot) and training environments (i.e., real-world and simulated). Im2Flow2Act comprises two components: a flow generation network and a flow-conditioned policy. The flow generation network, trained on human demonstration videos, generates object flow from the initial scene imag"},"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":"2407.15208","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2024-07-21T16:15:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ce68af869b4669e0bb732db3e6152cf4ef95bbbcf2387d0847c3cdfcaf433ed1","abstract_canon_sha256":"19ee46c60d4a9edfe0e81f292a7c903f88440129d05e0a5a16d35208fec5e081"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:35.144330Z","signature_b64":"FMpA64XykH7VzeAg6hQGPeb/k6p2ZmnVL5YvR/sn0irUp3iEPeijLZQFbJpnMmplwIEpkZSU8PP4VByPOM9mDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fc1c53abf5a056e8fac3a9a6f8ade48a0e1d5f210a20735fce28b87760ccb49c","last_reissued_at":"2026-07-05T09:15:35.143818Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:35.143818Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Flow as the Cross-Domain Manipulation Interface","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Cheng Chi, Gordon Wetzstein, Manuela Veloso, Mengda Xu, Shuran Song, Yinghao Xu, Zhenjia Xu","submitted_at":"2024-07-21T16:15:02Z","abstract_excerpt":"We present Im2Flow2Act, a scalable learning framework that enables robots to acquire real-world manipulation skills without the need of real-world robot training data. The key idea behind Im2Flow2Act is to use object flow as the manipulation interface, bridging domain gaps between different embodiments (i.e., human and robot) and training environments (i.e., real-world and simulated). Im2Flow2Act comprises two components: a flow generation network and a flow-conditioned policy. The flow generation network, trained on human demonstration videos, generates object flow from the initial scene imag"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.15208","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/2407.15208/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":"2407.15208","created_at":"2026-07-05T09:15:35.143877+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.15208v2","created_at":"2026-07-05T09:15:35.143877+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.15208","created_at":"2026-07-05T09:15:35.143877+00:00"},{"alias_kind":"pith_short_12","alias_value":"7QOFHK7VUBLO","created_at":"2026-07-05T09:15:35.143877+00:00"},{"alias_kind":"pith_short_16","alias_value":"7QOFHK7VUBLOR6WD","created_at":"2026-07-05T09:15:35.143877+00:00"},{"alias_kind":"pith_short_8","alias_value":"7QOFHK7V","created_at":"2026-07-05T09:15:35.143877+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":17,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08436","citing_title":"EgoWAM: World Action Models Beyond Pixels with In-the-Wild Egocentric Human Data","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2606.22113","citing_title":"KITE: Decoupling Kinematics and Interaction for Zero-Shot Cross-Embodiment Manipulation","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20781","citing_title":"World Action Models: A Survey","ref_index":177,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20135","citing_title":"Frequency-Aware Flow Matching for Continuous and Consistent Robotic Action Generation","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.16533","citing_title":"Kairos: A Regret-Aware Native World-Action Model Stack for Physical AI","ref_index":140,"is_internal_anchor":false},{"citing_arxiv_id":"2606.16917","citing_title":"Unified Motion-Action Modeling for Heterogeneous Robot Learning","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13515","citing_title":"MaskWAM: Unifying Mask Prompting and Prediction for World-Action Models","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2605.31234","citing_title":"HARP-VLA: Human-Robot Aligned Representation Learning for Vision-Language-Action Model","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29089","citing_title":"TAP-VLA: Tactile Annotation Prompting for Vision Language Action Models","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28813","citing_title":"Human2Any: Human-to-Robot Transfer via Constraint-Aware Compositional Planning","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2507.01099","citing_title":"Geometry-aware 4D Video Generation for Robot Manipulation","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2507.00990","citing_title":"Robotic Manipulation by Imitating Generated Videos Without Physical Demonstrations","ref_index":125,"is_internal_anchor":false},{"citing_arxiv_id":"2508.13998","citing_title":"Embodied-R1: Reinforced Embodied Reasoning for General Robotic Manipulation","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2510.01433","citing_title":"AFFORD2ACT: Affordance-Guided Automatic Keypoint Selection for Generalizable and Lightweight Robotic Manipulation","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12090","citing_title":"World Action Models: The Next Frontier in Embodied AI","ref_index":75,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12162","citing_title":"X-Imitator: Spatial-Aware Imitation Learning via Bidirectional Action-Pose Interaction","ref_index":67,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23249","citing_title":"BridgeACT: Bridging Human Demonstrations to Robot Actions via Unified Tool-Target Affordances","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7QOFHK7VUBLOR6WDVGTPRLPERI","json":"https://pith.science/pith/7QOFHK7VUBLOR6WDVGTPRLPERI.json","graph_json":"https://pith.science/api/pith-number/7QOFHK7VUBLOR6WDVGTPRLPERI/graph.json","events_json":"https://pith.science/api/pith-number/7QOFHK7VUBLOR6WDVGTPRLPERI/events.json","paper":"https://pith.science/paper/7QOFHK7V"},"agent_actions":{"view_html":"https://pith.science/pith/7QOFHK7VUBLOR6WDVGTPRLPERI","download_json":"https://pith.science/pith/7QOFHK7VUBLOR6WDVGTPRLPERI.json","view_paper":"https://pith.science/paper/7QOFHK7V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.15208&json=true","fetch_graph":"https://pith.science/api/pith-number/7QOFHK7VUBLOR6WDVGTPRLPERI/graph.json","fetch_events":"https://pith.science/api/pith-number/7QOFHK7VUBLOR6WDVGTPRLPERI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7QOFHK7VUBLOR6WDVGTPRLPERI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7QOFHK7VUBLOR6WDVGTPRLPERI/action/storage_attestation","attest_author":"https://pith.science/pith/7QOFHK7VUBLOR6WDVGTPRLPERI/action/author_attestation","sign_citation":"https://pith.science/pith/7QOFHK7VUBLOR6WDVGTPRLPERI/action/citation_signature","submit_replication":"https://pith.science/pith/7QOFHK7VUBLOR6WDVGTPRLPERI/action/replication_record"}},"created_at":"2026-07-05T09:15:35.143877+00:00","updated_at":"2026-07-05T09:15:35.143877+00:00"}