{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GQK53BPK6BXX5KH5R6DCTGIHV6","short_pith_number":"pith:GQK53BPK","schema_version":"1.0","canonical_sha256":"3415dd85eaf06f7ea8fd8f86299907afbefa440c42d7e0e10caed880200f58f0","source":{"kind":"arxiv","id":"2607.18236","version":1},"attestation_state":"computed","paper":{"title":"Patch Policy: Efficient Embodied Control via Dense Visual Representations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Ada Langford, Bowen Tan, Gaoyue Zhou, Lerrel Pinto, Yann LeCun, Zichen Jeff Cui","submitted_at":"2026-07-20T17:59:41Z","abstract_excerpt":"Pretrained dense visual features from Vision Transformers (ViTs) are powerful yet have been underutilized in robot learning. Modern robot policies either compress each observation into a single global token, or rely on visual backbones trained from scratch, sacrificing both fine-grained spatial detail and the benefits of large-scale visual pre-training. While there exist policies that do operate on dense patch features like large vision-language-action models (VLAs), they tend to be heavy and slow, inheriting the full cost of a billion-parameter vision-language model (VLM) backbone. We close t"},"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":"2607.18236","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-07-20T17:59:41Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1b30de5f5627615a0a2f545598b943b573fcd6d0207226f8d670490a2ad4379a","abstract_canon_sha256":"2b9b7c16fa840d2a618fa09e55d0999ac68e63ff44035f088d89af89eefdb512"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T02:22:29.462797Z","signature_b64":"n+TdH8R+NC8XNsfn6Sukj52TDOFzYSMKBL4+sRARfph4rdvAqgxE2Qxxez+C68AvIkbGDjbOG3E6BdedyG1iBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3415dd85eaf06f7ea8fd8f86299907afbefa440c42d7e0e10caed880200f58f0","last_reissued_at":"2026-07-21T02:22:29.461960Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T02:22:29.461960Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Patch Policy: Efficient Embodied Control via Dense Visual Representations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Ada Langford, Bowen Tan, Gaoyue Zhou, Lerrel Pinto, Yann LeCun, Zichen Jeff Cui","submitted_at":"2026-07-20T17:59:41Z","abstract_excerpt":"Pretrained dense visual features from Vision Transformers (ViTs) are powerful yet have been underutilized in robot learning. Modern robot policies either compress each observation into a single global token, or rely on visual backbones trained from scratch, sacrificing both fine-grained spatial detail and the benefits of large-scale visual pre-training. While there exist policies that do operate on dense patch features like large vision-language-action models (VLAs), they tend to be heavy and slow, inheriting the full cost of a billion-parameter vision-language model (VLM) backbone. We close t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.18236","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/2607.18236/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":"2607.18236","created_at":"2026-07-21T02:22:29.462391+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.18236v1","created_at":"2026-07-21T02:22:29.462391+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.18236","created_at":"2026-07-21T02:22:29.462391+00:00"},{"alias_kind":"pith_short_12","alias_value":"GQK53BPK6BXX","created_at":"2026-07-21T02:22:29.462391+00:00"},{"alias_kind":"pith_short_16","alias_value":"GQK53BPK6BXX5KH5","created_at":"2026-07-21T02:22:29.462391+00:00"},{"alias_kind":"pith_short_8","alias_value":"GQK53BPK","created_at":"2026-07-21T02:22:29.462391+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/GQK53BPK6BXX5KH5R6DCTGIHV6","json":"https://pith.science/pith/GQK53BPK6BXX5KH5R6DCTGIHV6.json","graph_json":"https://pith.science/api/pith-number/GQK53BPK6BXX5KH5R6DCTGIHV6/graph.json","events_json":"https://pith.science/api/pith-number/GQK53BPK6BXX5KH5R6DCTGIHV6/events.json","paper":"https://pith.science/paper/GQK53BPK"},"agent_actions":{"view_html":"https://pith.science/pith/GQK53BPK6BXX5KH5R6DCTGIHV6","download_json":"https://pith.science/pith/GQK53BPK6BXX5KH5R6DCTGIHV6.json","view_paper":"https://pith.science/paper/GQK53BPK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.18236&json=true","fetch_graph":"https://pith.science/api/pith-number/GQK53BPK6BXX5KH5R6DCTGIHV6/graph.json","fetch_events":"https://pith.science/api/pith-number/GQK53BPK6BXX5KH5R6DCTGIHV6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GQK53BPK6BXX5KH5R6DCTGIHV6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GQK53BPK6BXX5KH5R6DCTGIHV6/action/storage_attestation","attest_author":"https://pith.science/pith/GQK53BPK6BXX5KH5R6DCTGIHV6/action/author_attestation","sign_citation":"https://pith.science/pith/GQK53BPK6BXX5KH5R6DCTGIHV6/action/citation_signature","submit_replication":"https://pith.science/pith/GQK53BPK6BXX5KH5R6DCTGIHV6/action/replication_record"}},"created_at":"2026-07-21T02:22:29.462391+00:00","updated_at":"2026-07-21T02:22:29.462391+00:00"}