{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SHIBXQTQ4KWXAIOAU7WAZTPLDO","short_pith_number":"pith:SHIBXQTQ","schema_version":"1.0","canonical_sha256":"91d01bc270e2ad7021c0a7ec0ccdeb1bb997037fe8a7a79ac4f960116df7f94f","source":{"kind":"arxiv","id":"2410.19560","version":1},"attestation_state":"computed","paper":{"title":"Connecting Joint-Embedding Predictive Architecture with Contrastive Self-supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.IV","eess.SP"],"primary_cat":"cs.CV","authors_text":"Shengbang Tong, Shentong Mo","submitted_at":"2024-10-25T13:48:12Z","abstract_excerpt":"In recent advancements in unsupervised visual representation learning, the Joint-Embedding Predictive Architecture (JEPA) has emerged as a significant method for extracting visual features from unlabeled imagery through an innovative masking strategy. Despite its success, two primary limitations have been identified: the inefficacy of Exponential Moving Average (EMA) from I-JEPA in preventing entire collapse and the inadequacy of I-JEPA prediction in accurately learning the mean of patch representations. Addressing these challenges, this study introduces a novel framework, namely C-JEPA (Contr"},"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":"2410.19560","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-10-25T13:48:12Z","cross_cats_sorted":["cs.AI","cs.LG","eess.IV","eess.SP"],"title_canon_sha256":"47c659906b4a0a8175a0befa194aa4ff8742659d974a12c8be0829f9950f2602","abstract_canon_sha256":"e42624e40b8d09b102c19b95ac54f29dfc56fb57a913f4c79b479bdff730b476"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:25:58.273587Z","signature_b64":"jmcjq3hSv2P6yieyVn9K8fyNRud4oD1hSnRJtws+PkKF4OTlcnTXgjO27MQLvQSmuyH7RXEHOTFb4zuu7nlJCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"91d01bc270e2ad7021c0a7ec0ccdeb1bb997037fe8a7a79ac4f960116df7f94f","last_reissued_at":"2026-07-05T09:25:58.273120Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:25:58.273120Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Connecting Joint-Embedding Predictive Architecture with Contrastive Self-supervised Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.IV","eess.SP"],"primary_cat":"cs.CV","authors_text":"Shengbang Tong, Shentong Mo","submitted_at":"2024-10-25T13:48:12Z","abstract_excerpt":"In recent advancements in unsupervised visual representation learning, the Joint-Embedding Predictive Architecture (JEPA) has emerged as a significant method for extracting visual features from unlabeled imagery through an innovative masking strategy. Despite its success, two primary limitations have been identified: the inefficacy of Exponential Moving Average (EMA) from I-JEPA in preventing entire collapse and the inadequacy of I-JEPA prediction in accurately learning the mean of patch representations. Addressing these challenges, this study introduces a novel framework, namely C-JEPA (Contr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.19560","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/2410.19560/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":"2410.19560","created_at":"2026-07-05T09:25:58.273175+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.19560v1","created_at":"2026-07-05T09:25:58.273175+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.19560","created_at":"2026-07-05T09:25:58.273175+00:00"},{"alias_kind":"pith_short_12","alias_value":"SHIBXQTQ4KWX","created_at":"2026-07-05T09:25:58.273175+00:00"},{"alias_kind":"pith_short_16","alias_value":"SHIBXQTQ4KWXAIOA","created_at":"2026-07-05T09:25:58.273175+00:00"},{"alias_kind":"pith_short_8","alias_value":"SHIBXQTQ","created_at":"2026-07-05T09:25:58.273175+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.10385","citing_title":"Simplifying DINO via Coding Rate Regularization","ref_index":25,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SHIBXQTQ4KWXAIOAU7WAZTPLDO","json":"https://pith.science/pith/SHIBXQTQ4KWXAIOAU7WAZTPLDO.json","graph_json":"https://pith.science/api/pith-number/SHIBXQTQ4KWXAIOAU7WAZTPLDO/graph.json","events_json":"https://pith.science/api/pith-number/SHIBXQTQ4KWXAIOAU7WAZTPLDO/events.json","paper":"https://pith.science/paper/SHIBXQTQ"},"agent_actions":{"view_html":"https://pith.science/pith/SHIBXQTQ4KWXAIOAU7WAZTPLDO","download_json":"https://pith.science/pith/SHIBXQTQ4KWXAIOAU7WAZTPLDO.json","view_paper":"https://pith.science/paper/SHIBXQTQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.19560&json=true","fetch_graph":"https://pith.science/api/pith-number/SHIBXQTQ4KWXAIOAU7WAZTPLDO/graph.json","fetch_events":"https://pith.science/api/pith-number/SHIBXQTQ4KWXAIOAU7WAZTPLDO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SHIBXQTQ4KWXAIOAU7WAZTPLDO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SHIBXQTQ4KWXAIOAU7WAZTPLDO/action/storage_attestation","attest_author":"https://pith.science/pith/SHIBXQTQ4KWXAIOAU7WAZTPLDO/action/author_attestation","sign_citation":"https://pith.science/pith/SHIBXQTQ4KWXAIOAU7WAZTPLDO/action/citation_signature","submit_replication":"https://pith.science/pith/SHIBXQTQ4KWXAIOAU7WAZTPLDO/action/replication_record"}},"created_at":"2026-07-05T09:25:58.273175+00:00","updated_at":"2026-07-05T09:25:58.273175+00:00"}