{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:4WUZOX3TXD35INV2UEBSYVCB5Y","short_pith_number":"pith:4WUZOX3T","schema_version":"1.0","canonical_sha256":"e5a9975f73b8f7d436baa1032c5441ee3348f2ab7afcd63287baeece09499534","source":{"kind":"arxiv","id":"2310.18737","version":2},"attestation_state":"computed","paper":{"title":"Pre-training with Random Orthogonal Projection Image Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Maryam Haghighat, Peyman Moghadam, Piotr Koniusz, Shaheer Mohamed","submitted_at":"2023-10-28T15:42:07Z","abstract_excerpt":"Masked Image Modeling (MIM) is a powerful self-supervised strategy for visual pre-training without the use of labels. MIM applies random crops to input images, processes them with an encoder, and then recovers the masked inputs with a decoder, which encourages the network to capture and learn structural information about objects and scenes. The intermediate feature representations obtained from MIM are suitable for fine-tuning on downstream tasks. In this paper, we propose an Image Modeling framework based on random orthogonal projection instead of binary masking as in MIM. Our proposed Random"},"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":"2310.18737","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-10-28T15:42:07Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"2644103e89808cc2fd21ae8d1b65f24be58292ed361b9012b007d05422fcf77f","abstract_canon_sha256":"538a28e855a01a5b359283bda127370c8f984feebacf4c0c519893bb1240531f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:10:19.176362Z","signature_b64":"GpvhXCnaiwsIW2AQFDm9PaHFCYsG7ZxtjLjIEiyFl7fUEXo/vhDbA21bIM/H8FqvXTt08QCPk6nR5GHdnzLkBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e5a9975f73b8f7d436baa1032c5441ee3348f2ab7afcd63287baeece09499534","last_reissued_at":"2026-07-05T08:10:19.175877Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:10:19.175877Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pre-training with Random Orthogonal Projection Image Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Maryam Haghighat, Peyman Moghadam, Piotr Koniusz, Shaheer Mohamed","submitted_at":"2023-10-28T15:42:07Z","abstract_excerpt":"Masked Image Modeling (MIM) is a powerful self-supervised strategy for visual pre-training without the use of labels. MIM applies random crops to input images, processes them with an encoder, and then recovers the masked inputs with a decoder, which encourages the network to capture and learn structural information about objects and scenes. The intermediate feature representations obtained from MIM are suitable for fine-tuning on downstream tasks. In this paper, we propose an Image Modeling framework based on random orthogonal projection instead of binary masking as in MIM. Our proposed Random"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.18737","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/2310.18737/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":"2310.18737","created_at":"2026-07-05T08:10:19.175934+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.18737v2","created_at":"2026-07-05T08:10:19.175934+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.18737","created_at":"2026-07-05T08:10:19.175934+00:00"},{"alias_kind":"pith_short_12","alias_value":"4WUZOX3TXD35","created_at":"2026-07-05T08:10:19.175934+00:00"},{"alias_kind":"pith_short_16","alias_value":"4WUZOX3TXD35INV2","created_at":"2026-07-05T08:10:19.175934+00:00"},{"alias_kind":"pith_short_8","alias_value":"4WUZOX3T","created_at":"2026-07-05T08:10:19.175934+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.14044","citing_title":"Generalized Category Discovery via Token Manifold Capacity Learning","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4WUZOX3TXD35INV2UEBSYVCB5Y","json":"https://pith.science/pith/4WUZOX3TXD35INV2UEBSYVCB5Y.json","graph_json":"https://pith.science/api/pith-number/4WUZOX3TXD35INV2UEBSYVCB5Y/graph.json","events_json":"https://pith.science/api/pith-number/4WUZOX3TXD35INV2UEBSYVCB5Y/events.json","paper":"https://pith.science/paper/4WUZOX3T"},"agent_actions":{"view_html":"https://pith.science/pith/4WUZOX3TXD35INV2UEBSYVCB5Y","download_json":"https://pith.science/pith/4WUZOX3TXD35INV2UEBSYVCB5Y.json","view_paper":"https://pith.science/paper/4WUZOX3T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.18737&json=true","fetch_graph":"https://pith.science/api/pith-number/4WUZOX3TXD35INV2UEBSYVCB5Y/graph.json","fetch_events":"https://pith.science/api/pith-number/4WUZOX3TXD35INV2UEBSYVCB5Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4WUZOX3TXD35INV2UEBSYVCB5Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4WUZOX3TXD35INV2UEBSYVCB5Y/action/storage_attestation","attest_author":"https://pith.science/pith/4WUZOX3TXD35INV2UEBSYVCB5Y/action/author_attestation","sign_citation":"https://pith.science/pith/4WUZOX3TXD35INV2UEBSYVCB5Y/action/citation_signature","submit_replication":"https://pith.science/pith/4WUZOX3TXD35INV2UEBSYVCB5Y/action/replication_record"}},"created_at":"2026-07-05T08:10:19.175934+00:00","updated_at":"2026-07-05T08:10:19.175934+00:00"}