{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PQ4U6KAE3TW4CLO3UG4JPFO5OW","short_pith_number":"pith:PQ4U6KAE","schema_version":"1.0","canonical_sha256":"7c394f2804dcedc12ddba1b89795dd758d467c0adf8e6e4d4aae5522d8f2cbc8","source":{"kind":"arxiv","id":"2607.11167","version":1},"attestation_state":"computed","paper":{"title":"Pix2Act: Image-Space Manipulation Policies with Equivariant Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Boce Hu, Dian Wang, Fangzhou Lin, Haojie Huang, Haotian Liu, Linfeng Zhao, Mingxi Jia, Robert Platt, Robin Walters, Si-Yuan Huang, Yu Qi, Zhang Ye","submitted_at":"2026-07-13T07:03:29Z","abstract_excerpt":"Representing manipulation actions as 2D trajectories in the camera plane provides a compact and interpretable basis for learning complex 3D manipulation policies. However, it also creates challenges from out-of-frame trajectories and limited precision. We propose Pix2Act, an imitation learning method that addresses these challenges by generating continuous image-space keypoint trajectories in each camera plane and losslessly recovering end-effector poses via triangulation. This reformulates high-dimensional 3D control as a simpler, more learnable 2D prediction problem. Crucially, it aligns obs"},"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":"2607.11167","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-07-13T07:03:29Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"802897b650b8334be4ce36c3f3c5bd0770d01dbcfe36821cc953ef0429217584","abstract_canon_sha256":"283a57186d3f1c095cc8ab4d2c5ec17fd79d229471ae806a0dc4fb9fe85235c6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T01:22:20.000904Z","signature_b64":"IN6s9wxPsKb4tdf+omIR06hGCRZUd3enIwtDm7G6FIrHuW35PJ0yb1RJ5IXlUxw/mcWEsJX2/QvAL6eEBE+FBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7c394f2804dcedc12ddba1b89795dd758d467c0adf8e6e4d4aae5522d8f2cbc8","last_reissued_at":"2026-07-14T01:22:20.000102Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T01:22:20.000102Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pix2Act: Image-Space Manipulation Policies with Equivariant Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Boce Hu, Dian Wang, Fangzhou Lin, Haojie Huang, Haotian Liu, Linfeng Zhao, Mingxi Jia, Robert Platt, Robin Walters, Si-Yuan Huang, Yu Qi, Zhang Ye","submitted_at":"2026-07-13T07:03:29Z","abstract_excerpt":"Representing manipulation actions as 2D trajectories in the camera plane provides a compact and interpretable basis for learning complex 3D manipulation policies. However, it also creates challenges from out-of-frame trajectories and limited precision. We propose Pix2Act, an imitation learning method that addresses these challenges by generating continuous image-space keypoint trajectories in each camera plane and losslessly recovering end-effector poses via triangulation. This reformulates high-dimensional 3D control as a simpler, more learnable 2D prediction problem. Crucially, it aligns obs"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.11167","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/2607.11167/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":"2607.11167","created_at":"2026-07-14T01:22:20.000520+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.11167v1","created_at":"2026-07-14T01:22:20.000520+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.11167","created_at":"2026-07-14T01:22:20.000520+00:00"},{"alias_kind":"pith_short_12","alias_value":"PQ4U6KAE3TW4","created_at":"2026-07-14T01:22:20.000520+00:00"},{"alias_kind":"pith_short_16","alias_value":"PQ4U6KAE3TW4CLO3","created_at":"2026-07-14T01:22:20.000520+00:00"},{"alias_kind":"pith_short_8","alias_value":"PQ4U6KAE","created_at":"2026-07-14T01:22:20.000520+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/PQ4U6KAE3TW4CLO3UG4JPFO5OW","json":"https://pith.science/pith/PQ4U6KAE3TW4CLO3UG4JPFO5OW.json","graph_json":"https://pith.science/api/pith-number/PQ4U6KAE3TW4CLO3UG4JPFO5OW/graph.json","events_json":"https://pith.science/api/pith-number/PQ4U6KAE3TW4CLO3UG4JPFO5OW/events.json","paper":"https://pith.science/paper/PQ4U6KAE"},"agent_actions":{"view_html":"https://pith.science/pith/PQ4U6KAE3TW4CLO3UG4JPFO5OW","download_json":"https://pith.science/pith/PQ4U6KAE3TW4CLO3UG4JPFO5OW.json","view_paper":"https://pith.science/paper/PQ4U6KAE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.11167&json=true","fetch_graph":"https://pith.science/api/pith-number/PQ4U6KAE3TW4CLO3UG4JPFO5OW/graph.json","fetch_events":"https://pith.science/api/pith-number/PQ4U6KAE3TW4CLO3UG4JPFO5OW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PQ4U6KAE3TW4CLO3UG4JPFO5OW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PQ4U6KAE3TW4CLO3UG4JPFO5OW/action/storage_attestation","attest_author":"https://pith.science/pith/PQ4U6KAE3TW4CLO3UG4JPFO5OW/action/author_attestation","sign_citation":"https://pith.science/pith/PQ4U6KAE3TW4CLO3UG4JPFO5OW/action/citation_signature","submit_replication":"https://pith.science/pith/PQ4U6KAE3TW4CLO3UG4JPFO5OW/action/replication_record"}},"created_at":"2026-07-14T01:22:20.000520+00:00","updated_at":"2026-07-14T01:22:20.000520+00:00"}