{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WIDBK3J3ZHBZAOJYU7ALNN37SZ","short_pith_number":"pith:WIDBK3J3","schema_version":"1.0","canonical_sha256":"b206156d3bc9c3903938a7c0b6b77f964c2fa4bf51417f193f635644fce2126e","source":{"kind":"arxiv","id":"2505.07817","version":2},"attestation_state":"computed","paper":{"title":"Pixel Motion as Universal Representation for Robot Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Cristina Mata, E-Ro Nguyen, Jongwoo Park, Kanchana Ranasinghe, Michael S Ryoo, Xiang Li","submitted_at":"2025-05-12T17:59:32Z","abstract_excerpt":"We present LangToMo, a vision-language-action framework structured as a dual-system architecture that uses pixel motion forecasts as intermediate representations. Our high-level System 2, an image diffusion model, generates text-conditioned pixel motion sequences from a single frame to guide robot control. Pixel motion-a universal, interpretable, and motion-centric representation-can be extracted from videos in a weakly-supervised manner, enabling diffusion model training on any video-caption data. Treating generated pixel motion as learned universal representations, our low level System 1 mod"},"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":"2505.07817","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-05-12T17:59:32Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"0129056e61cc7d42ae795c4cb6cdbc14ceffa5c05dc3ac7383801d23bd257e51","abstract_canon_sha256":"6d20effd2c92ba5be982825c8df0441cc17fc986647455e10cea7e2630305c35"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:00:49.904964Z","signature_b64":"ho8UclpNROrlDgRmMmn2vJzpkIHN/wG56xd7Tl5cQnVt4uq8DBcp1Evr1JJcWel6XQUU907QIazahLTrYZn1DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b206156d3bc9c3903938a7c0b6b77f964c2fa4bf51417f193f635644fce2126e","last_reissued_at":"2026-07-05T12:00:49.904418Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:00:49.904418Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pixel Motion as Universal Representation for Robot Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.RO","authors_text":"Cristina Mata, E-Ro Nguyen, Jongwoo Park, Kanchana Ranasinghe, Michael S Ryoo, Xiang Li","submitted_at":"2025-05-12T17:59:32Z","abstract_excerpt":"We present LangToMo, a vision-language-action framework structured as a dual-system architecture that uses pixel motion forecasts as intermediate representations. Our high-level System 2, an image diffusion model, generates text-conditioned pixel motion sequences from a single frame to guide robot control. Pixel motion-a universal, interpretable, and motion-centric representation-can be extracted from videos in a weakly-supervised manner, enabling diffusion model training on any video-caption data. Treating generated pixel motion as learned universal representations, our low level System 1 mod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.07817","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/2505.07817/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":"2505.07817","created_at":"2026-07-05T12:00:49.904485+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.07817v2","created_at":"2026-07-05T12:00:49.904485+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.07817","created_at":"2026-07-05T12:00:49.904485+00:00"},{"alias_kind":"pith_short_12","alias_value":"WIDBK3J3ZHBZ","created_at":"2026-07-05T12:00:49.904485+00:00"},{"alias_kind":"pith_short_16","alias_value":"WIDBK3J3ZHBZAOJY","created_at":"2026-07-05T12:00:49.904485+00:00"},{"alias_kind":"pith_short_8","alias_value":"WIDBK3J3","created_at":"2026-07-05T12:00:49.904485+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.25829","citing_title":"OASIS: Observation-Action Space Alignment via SE(3) Trajectory Prediction for Robotic Manipulation","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23856","citing_title":"Point Tracking Improves World Action Models","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2507.04447","citing_title":"DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge","ref_index":51,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WIDBK3J3ZHBZAOJYU7ALNN37SZ","json":"https://pith.science/pith/WIDBK3J3ZHBZAOJYU7ALNN37SZ.json","graph_json":"https://pith.science/api/pith-number/WIDBK3J3ZHBZAOJYU7ALNN37SZ/graph.json","events_json":"https://pith.science/api/pith-number/WIDBK3J3ZHBZAOJYU7ALNN37SZ/events.json","paper":"https://pith.science/paper/WIDBK3J3"},"agent_actions":{"view_html":"https://pith.science/pith/WIDBK3J3ZHBZAOJYU7ALNN37SZ","download_json":"https://pith.science/pith/WIDBK3J3ZHBZAOJYU7ALNN37SZ.json","view_paper":"https://pith.science/paper/WIDBK3J3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.07817&json=true","fetch_graph":"https://pith.science/api/pith-number/WIDBK3J3ZHBZAOJYU7ALNN37SZ/graph.json","fetch_events":"https://pith.science/api/pith-number/WIDBK3J3ZHBZAOJYU7ALNN37SZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WIDBK3J3ZHBZAOJYU7ALNN37SZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WIDBK3J3ZHBZAOJYU7ALNN37SZ/action/storage_attestation","attest_author":"https://pith.science/pith/WIDBK3J3ZHBZAOJYU7ALNN37SZ/action/author_attestation","sign_citation":"https://pith.science/pith/WIDBK3J3ZHBZAOJYU7ALNN37SZ/action/citation_signature","submit_replication":"https://pith.science/pith/WIDBK3J3ZHBZAOJYU7ALNN37SZ/action/replication_record"}},"created_at":"2026-07-05T12:00:49.904485+00:00","updated_at":"2026-07-05T12:00:49.904485+00:00"}