{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:APTM2RGTNNX6WQYUQGHRDZFNZF","short_pith_number":"pith:APTM2RGT","schema_version":"1.0","canonical_sha256":"03e6cd44d36b6feb4314818f11e4adc97aef8ffe8273958dd4f6eb7db107a00c","source":{"kind":"arxiv","id":"2507.12440","version":3},"attestation_state":"computed","paper":{"title":"EgoVLA: Learning Vision-Language-Action Models from Egocentric Human Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.RO","authors_text":"An-Chieh Cheng, BoRui Li, Hongxu Yin, Qinxi Yu, Ri-Zhao Qiu, Ruihan Yang, Rui Yan, Sifei Liu, Song Han, Xiaolong Wang, Xueyan Zou, Xuxin Cheng, Yao Lu, Yecheng Wu, Yunhao Fang","submitted_at":"2025-07-16T17:27:44Z","abstract_excerpt":"Real robot data collection for imitation learning has led to significant advancements in robotic manipulation. However, the requirement for robot hardware in the process fundamentally constrains the scale of the data. In this paper, we explore training Vision-Language-Action (VLA) models using egocentric human videos. The benefit of using human videos is not only for their scale but more importantly for the richness of scenes and tasks. With a VLA trained on human video that predicts human wrist and hand actions, we can perform Inverse Kinematics and retargeting to convert the human actions to"},"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":"2507.12440","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-07-16T17:27:44Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"title_canon_sha256":"35d6c8a08ed7e862faa491421fdea9b3a680d66ca768d255f737e8d6fcc54559","abstract_canon_sha256":"f9e2b4dcdf8182edd04a77c1e3edb2b0a086f24bd68c6138699b57a11247243b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T04:27:34.074456Z","signature_b64":"pUUc1EWri/w4sNpwV7MgbN+tvv7f8zKaqB/Ceo37Hi/RljOLVx1ZfUA9HOfSCGBS/hyUhlFUExJuuOs0ddKDBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"03e6cd44d36b6feb4314818f11e4adc97aef8ffe8273958dd4f6eb7db107a00c","last_reissued_at":"2026-05-21T04:27:34.072705Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T04:27:34.072705Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EgoVLA: Learning Vision-Language-Action Models from Egocentric Human Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.RO","authors_text":"An-Chieh Cheng, BoRui Li, Hongxu Yin, Qinxi Yu, Ri-Zhao Qiu, Ruihan Yang, Rui Yan, Sifei Liu, Song Han, Xiaolong Wang, Xueyan Zou, Xuxin Cheng, Yao Lu, Yecheng Wu, Yunhao Fang","submitted_at":"2025-07-16T17:27:44Z","abstract_excerpt":"Real robot data collection for imitation learning has led to significant advancements in robotic manipulation. However, the requirement for robot hardware in the process fundamentally constrains the scale of the data. In this paper, we explore training Vision-Language-Action (VLA) models using egocentric human videos. The benefit of using human videos is not only for their scale but more importantly for the richness of scenes and tasks. With a VLA trained on human video that predicts human wrist and hand actions, we can perform Inverse Kinematics and retargeting to convert the human actions to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.12440","kind":"arxiv","version":3},"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/2507.12440/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":"2507.12440","created_at":"2026-05-21T04:27:34.072779+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.12440v3","created_at":"2026-05-21T04:27:34.072779+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.12440","created_at":"2026-05-21T04:27:34.072779+00:00"},{"alias_kind":"pith_short_12","alias_value":"APTM2RGTNNX6","created_at":"2026-05-21T04:27:34.072779+00:00"},{"alias_kind":"pith_short_16","alias_value":"APTM2RGTNNX6WQYU","created_at":"2026-05-21T04:27:34.072779+00:00"},{"alias_kind":"pith_short_8","alias_value":"APTM2RGT","created_at":"2026-05-21T04:27:34.072779+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":23,"internal_anchor_count":23,"sample":[{"citing_arxiv_id":"2604.24681","citing_title":"Learning Human-Intention Priors from Large-Scale Human Demonstrations for Robotic Manipulation","ref_index":42,"is_internal_anchor":true},{"citing_arxiv_id":"2603.03243","citing_title":"HoMMI: Learning Whole-Body Mobile Manipulation from Human Demonstrations","ref_index":42,"is_internal_anchor":true},{"citing_arxiv_id":"2604.07993","citing_title":"HEX: Humanoid-Aligned Experts for Cross-Embodiment Whole-Body Manipulation","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21133","citing_title":"Humanoid Whole-Body Manipulation via Active Spatial Brain and Generalizable Action Cerebellum","ref_index":47,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16412","citing_title":"SCAR: Self-Supervised Continuous Action Representation Learning","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18553","citing_title":"StableHand: Quality-Aware Flow Matching for World-Space Dual-Hand Motion Estimation from Egocentric Video","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18722","citing_title":"Dexora: Open-source VLA for High-DoF Bimanual Dexterity","ref_index":31,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16797","citing_title":"EgoKit: Towards Unified Low-Cost Egocentric Data Collection with Heterogeneous Devices","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2602.06949","citing_title":"DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos","ref_index":111,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12038","citing_title":"OmniHumanoid: Streaming Cross-Embodiment Video Generation with Paired-Free Adaptation","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12090","citing_title":"World Action Models: The Next Frontier in Embodied AI","ref_index":204,"is_internal_anchor":true},{"citing_arxiv_id":"2604.24681","citing_title":"Learning Human-Intention Priors from Large-Scale Human Demonstrations for Robotic Manipulation","ref_index":42,"is_internal_anchor":true},{"citing_arxiv_id":"2604.23570","citing_title":"EgoLive: A Large-Scale Egocentric Dataset from Real-World Human Tasks","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22615","citing_title":"GazeVLA: Learning Human Intention for Robotic Manipulation","ref_index":72,"is_internal_anchor":true},{"citing_arxiv_id":"2605.00078","citing_title":"Being-H0.7: A Latent World-Action Model from Egocentric Videos","ref_index":106,"is_internal_anchor":true},{"citing_arxiv_id":"2604.21241","citing_title":"CorridorVLA: Explicit Spatial Constraints for Generative Action Heads via Sparse Anchors","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2604.19734","citing_title":"UniT: Toward a Unified Physical Language for Human-to-Humanoid Policy Learning and World Modeling","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2604.10677","citing_title":"LIDEA: Human-to-Robot Imitation Learning via Implicit Feature Distillation and Explicit Geometry Alignment","ref_index":21,"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":55,"is_internal_anchor":true},{"citing_arxiv_id":"2604.07517","citing_title":"Grasp as You Dream: Imitating Functional Grasping from Generated Human Demonstrations","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2604.08534","citing_title":"ActiveGlasses: Learning Manipulation with Active Vision from Ego-centric Human Demonstration","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2604.07993","citing_title":"HEX: Humanoid-Aligned Experts for Cross-Embodiment Whole-Body Manipulation","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2604.15483","citing_title":"${\\pi}_{0.7}$: a Steerable Generalist Robotic Foundation Model with Emergent Capabilities","ref_index":48,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/APTM2RGTNNX6WQYUQGHRDZFNZF","json":"https://pith.science/pith/APTM2RGTNNX6WQYUQGHRDZFNZF.json","graph_json":"https://pith.science/api/pith-number/APTM2RGTNNX6WQYUQGHRDZFNZF/graph.json","events_json":"https://pith.science/api/pith-number/APTM2RGTNNX6WQYUQGHRDZFNZF/events.json","paper":"https://pith.science/paper/APTM2RGT"},"agent_actions":{"view_html":"https://pith.science/pith/APTM2RGTNNX6WQYUQGHRDZFNZF","download_json":"https://pith.science/pith/APTM2RGTNNX6WQYUQGHRDZFNZF.json","view_paper":"https://pith.science/paper/APTM2RGT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.12440&json=true","fetch_graph":"https://pith.science/api/pith-number/APTM2RGTNNX6WQYUQGHRDZFNZF/graph.json","fetch_events":"https://pith.science/api/pith-number/APTM2RGTNNX6WQYUQGHRDZFNZF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/APTM2RGTNNX6WQYUQGHRDZFNZF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/APTM2RGTNNX6WQYUQGHRDZFNZF/action/storage_attestation","attest_author":"https://pith.science/pith/APTM2RGTNNX6WQYUQGHRDZFNZF/action/author_attestation","sign_citation":"https://pith.science/pith/APTM2RGTNNX6WQYUQGHRDZFNZF/action/citation_signature","submit_replication":"https://pith.science/pith/APTM2RGTNNX6WQYUQGHRDZFNZF/action/replication_record"}},"created_at":"2026-05-21T04:27:34.072779+00:00","updated_at":"2026-05-21T04:27:34.072779+00:00"}