{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2CXEBZWBOJA2NH5RV5PFN57DMU","short_pith_number":"pith:2CXEBZWB","schema_version":"1.0","canonical_sha256":"d0ae40e6c17241a69fb1af5e56f7e36500094a6b1f5c6a4b68347281f941549a","source":{"kind":"arxiv","id":"2502.09140","version":1},"attestation_state":"computed","paper":{"title":"Replay-free Online Continual Learning with Self-Supervised MultiPatches","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Alex Gomez-Villa, Andrea Cossu, Antonio Carta, Giacomo Cignoni, Joost Van De Weijer","submitted_at":"2025-02-13T10:15:16Z","abstract_excerpt":"Online Continual Learning (OCL) methods train a model on a non-stationary data stream where only a few examples are available at a time, often leveraging replay strategies. However, usage of replay is sometimes forbidden, especially in applications with strict privacy regulations. Therefore, we propose Continual MultiPatches (CMP), an effective plug-in for existing OCL self-supervised learning strategies that avoids the use of replay samples. CMP generates multiple patches from a single example and projects them into a shared feature space, where patches coming from the same example are pushed"},"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":"2502.09140","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T10:15:16Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"a2209b903875c61aca3e59caca924a030f02fffb9e84147f059b8d46faf05999","abstract_canon_sha256":"86e2875656c498ae4c73e507ed01357dcbfccd3116fc5fad19e85f66cdc6bf77"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:52.893848Z","signature_b64":"GdBr+LfcRR2/wpKi+jkYCVB4WHyprp4Yj6o2s7MMLOJJYo0JQX/g+yNEfPX5BuaYxtLERPYWO7TIUIs3BlLWDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d0ae40e6c17241a69fb1af5e56f7e36500094a6b1f5c6a4b68347281f941549a","last_reissued_at":"2026-07-05T10:13:52.893342Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:52.893342Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Replay-free Online Continual Learning with Self-Supervised MultiPatches","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Alex Gomez-Villa, Andrea Cossu, Antonio Carta, Giacomo Cignoni, Joost Van De Weijer","submitted_at":"2025-02-13T10:15:16Z","abstract_excerpt":"Online Continual Learning (OCL) methods train a model on a non-stationary data stream where only a few examples are available at a time, often leveraging replay strategies. However, usage of replay is sometimes forbidden, especially in applications with strict privacy regulations. Therefore, we propose Continual MultiPatches (CMP), an effective plug-in for existing OCL self-supervised learning strategies that avoids the use of replay samples. CMP generates multiple patches from a single example and projects them into a shared feature space, where patches coming from the same example are pushed"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.09140","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/2502.09140/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":"2502.09140","created_at":"2026-07-05T10:13:52.893402+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.09140v1","created_at":"2026-07-05T10:13:52.893402+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.09140","created_at":"2026-07-05T10:13:52.893402+00:00"},{"alias_kind":"pith_short_12","alias_value":"2CXEBZWBOJA2","created_at":"2026-07-05T10:13:52.893402+00:00"},{"alias_kind":"pith_short_16","alias_value":"2CXEBZWBOJA2NH5R","created_at":"2026-07-05T10:13:52.893402+00:00"},{"alias_kind":"pith_short_8","alias_value":"2CXEBZWB","created_at":"2026-07-05T10:13:52.893402+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.10586","citing_title":"Preventing Latent Rehearsal Decay in Online Continual SSL with SOLAR","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2CXEBZWBOJA2NH5RV5PFN57DMU","json":"https://pith.science/pith/2CXEBZWBOJA2NH5RV5PFN57DMU.json","graph_json":"https://pith.science/api/pith-number/2CXEBZWBOJA2NH5RV5PFN57DMU/graph.json","events_json":"https://pith.science/api/pith-number/2CXEBZWBOJA2NH5RV5PFN57DMU/events.json","paper":"https://pith.science/paper/2CXEBZWB"},"agent_actions":{"view_html":"https://pith.science/pith/2CXEBZWBOJA2NH5RV5PFN57DMU","download_json":"https://pith.science/pith/2CXEBZWBOJA2NH5RV5PFN57DMU.json","view_paper":"https://pith.science/paper/2CXEBZWB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.09140&json=true","fetch_graph":"https://pith.science/api/pith-number/2CXEBZWBOJA2NH5RV5PFN57DMU/graph.json","fetch_events":"https://pith.science/api/pith-number/2CXEBZWBOJA2NH5RV5PFN57DMU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2CXEBZWBOJA2NH5RV5PFN57DMU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2CXEBZWBOJA2NH5RV5PFN57DMU/action/storage_attestation","attest_author":"https://pith.science/pith/2CXEBZWBOJA2NH5RV5PFN57DMU/action/author_attestation","sign_citation":"https://pith.science/pith/2CXEBZWBOJA2NH5RV5PFN57DMU/action/citation_signature","submit_replication":"https://pith.science/pith/2CXEBZWBOJA2NH5RV5PFN57DMU/action/replication_record"}},"created_at":"2026-07-05T10:13:52.893402+00:00","updated_at":"2026-07-05T10:13:52.893402+00:00"}