{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MGLKRT42H6RYOLRMDKWQQO3E5S","short_pith_number":"pith:MGLKRT42","schema_version":"1.0","canonical_sha256":"6196a8cf9a3fa3872e2c1aad083b64ec99e4cd2b01fa356f69f16761389e16f1","source":{"kind":"arxiv","id":"2408.11054","version":3},"attestation_state":"computed","paper":{"title":"Near, far: Patch-ordering enhances vision foundation models' scene understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Francesco Locatello, Gertjan Burghouts, Mohammadreza Salehi, Valentinos Pariza, Yuki M. Asano","submitted_at":"2024-08-20T17:58:59Z","abstract_excerpt":"We introduce NeCo: Patch Neighbor Consistency, a novel self-supervised training loss that enforces patch-level nearest neighbor consistency across a student and teacher model. Compared to contrastive approaches that only yield binary learning signals, i.e., 'attract' and 'repel', this approach benefits from the more fine-grained learning signal of sorting spatially dense features relative to reference patches. Our method leverages differentiable sorting applied on top of pretrained representations, such as DINOv2-registers to bootstrap the learning signal and further improve upon them. This de"},"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":"2408.11054","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-08-20T17:58:59Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0d18221597609970bfc1e51b796da46f746cd9182fdbff9e4830d35d3b842259","abstract_canon_sha256":"97e10342396ff4a3dfed4fcba5a2f7a1498117a62581fbb018b1239eefb46401"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:50:09.026095Z","signature_b64":"FIes+7o5A9mltAbzmlgxYH3IvOxwc8f7u6yFcvCwgO60SOcnpgoP2Ay62yQCQ0x7LHSrDXETpa0n3241RtijBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6196a8cf9a3fa3872e2c1aad083b64ec99e4cd2b01fa356f69f16761389e16f1","last_reissued_at":"2026-07-05T10:50:09.025670Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:50:09.025670Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Near, far: Patch-ordering enhances vision foundation models' scene understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Francesco Locatello, Gertjan Burghouts, Mohammadreza Salehi, Valentinos Pariza, Yuki M. Asano","submitted_at":"2024-08-20T17:58:59Z","abstract_excerpt":"We introduce NeCo: Patch Neighbor Consistency, a novel self-supervised training loss that enforces patch-level nearest neighbor consistency across a student and teacher model. Compared to contrastive approaches that only yield binary learning signals, i.e., 'attract' and 'repel', this approach benefits from the more fine-grained learning signal of sorting spatially dense features relative to reference patches. Our method leverages differentiable sorting applied on top of pretrained representations, such as DINOv2-registers to bootstrap the learning signal and further improve upon them. This de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.11054","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/2408.11054/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":"2408.11054","created_at":"2026-07-05T10:50:09.025725+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.11054v3","created_at":"2026-07-05T10:50:09.025725+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.11054","created_at":"2026-07-05T10:50:09.025725+00:00"},{"alias_kind":"pith_short_12","alias_value":"MGLKRT42H6RY","created_at":"2026-07-05T10:50:09.025725+00:00"},{"alias_kind":"pith_short_16","alias_value":"MGLKRT42H6RYOLRM","created_at":"2026-07-05T10:50:09.025725+00:00"},{"alias_kind":"pith_short_8","alias_value":"MGLKRT42","created_at":"2026-07-05T10:50:09.025725+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.09963","citing_title":"Learning to Perceive \"Where\": Spatial Pretext Tasks for Robust Self-Supervised Learning","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MGLKRT42H6RYOLRMDKWQQO3E5S","json":"https://pith.science/pith/MGLKRT42H6RYOLRMDKWQQO3E5S.json","graph_json":"https://pith.science/api/pith-number/MGLKRT42H6RYOLRMDKWQQO3E5S/graph.json","events_json":"https://pith.science/api/pith-number/MGLKRT42H6RYOLRMDKWQQO3E5S/events.json","paper":"https://pith.science/paper/MGLKRT42"},"agent_actions":{"view_html":"https://pith.science/pith/MGLKRT42H6RYOLRMDKWQQO3E5S","download_json":"https://pith.science/pith/MGLKRT42H6RYOLRMDKWQQO3E5S.json","view_paper":"https://pith.science/paper/MGLKRT42","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.11054&json=true","fetch_graph":"https://pith.science/api/pith-number/MGLKRT42H6RYOLRMDKWQQO3E5S/graph.json","fetch_events":"https://pith.science/api/pith-number/MGLKRT42H6RYOLRMDKWQQO3E5S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MGLKRT42H6RYOLRMDKWQQO3E5S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MGLKRT42H6RYOLRMDKWQQO3E5S/action/storage_attestation","attest_author":"https://pith.science/pith/MGLKRT42H6RYOLRMDKWQQO3E5S/action/author_attestation","sign_citation":"https://pith.science/pith/MGLKRT42H6RYOLRMDKWQQO3E5S/action/citation_signature","submit_replication":"https://pith.science/pith/MGLKRT42H6RYOLRMDKWQQO3E5S/action/replication_record"}},"created_at":"2026-07-05T10:50:09.025725+00:00","updated_at":"2026-07-05T10:50:09.025725+00:00"}