{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:O2A4P2WD75O4QWFN22GMG6ICZQ","short_pith_number":"pith:O2A4P2WD","schema_version":"1.0","canonical_sha256":"7681c7eac3ff5dc858add68cc37902cc14355cfe19b903cdd435dac24aab3035","source":{"kind":"arxiv","id":"2405.17995","version":1},"attestation_state":"computed","paper":{"title":"DMT-JEPA: Discriminative Masked Targets for Joint-Embedding Predictive Architecture","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Shentong Mo, Sukmin Yun","submitted_at":"2024-05-28T09:28:52Z","abstract_excerpt":"The joint-embedding predictive architecture (JEPA) recently has shown impressive results in extracting visual representations from unlabeled imagery under a masking strategy. However, we reveal its disadvantages, notably its insufficient understanding of local semantics. This deficiency originates from masked modeling in the embedding space, resulting in a reduction of discriminative power and can even lead to the neglect of critical local semantics. To bridge this gap, we introduce DMT-JEPA, a novel masked modeling objective rooted in JEPA, specifically designed to generate discriminative lat"},"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":"2405.17995","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-05-28T09:28:52Z","cross_cats_sorted":["cs.AI","cs.LG","eess.IV"],"title_canon_sha256":"8418d4474d9596ca8346b80c3a5463cea492393540e2374bb3a34ca2143f6d31","abstract_canon_sha256":"3405aaa21dedd92bb36dabb0b1c0ca87601307d01eb6a01a2c0f381471d77e95"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:24:14.307648Z","signature_b64":"tegMIDWiZLKLlBeabFbgPl1dINoN+/p/6wqG/MsdXwP8TRxhjp6yvqYavIfXRkUYkN3QEFbK+pjbmshF/qnhBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7681c7eac3ff5dc858add68cc37902cc14355cfe19b903cdd435dac24aab3035","last_reissued_at":"2026-07-05T08:24:14.307220Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:24:14.307220Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DMT-JEPA: Discriminative Masked Targets for Joint-Embedding Predictive Architecture","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Shentong Mo, Sukmin Yun","submitted_at":"2024-05-28T09:28:52Z","abstract_excerpt":"The joint-embedding predictive architecture (JEPA) recently has shown impressive results in extracting visual representations from unlabeled imagery under a masking strategy. However, we reveal its disadvantages, notably its insufficient understanding of local semantics. This deficiency originates from masked modeling in the embedding space, resulting in a reduction of discriminative power and can even lead to the neglect of critical local semantics. To bridge this gap, we introduce DMT-JEPA, a novel masked modeling objective rooted in JEPA, specifically designed to generate discriminative lat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.17995","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/2405.17995/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":"2405.17995","created_at":"2026-07-05T08:24:14.307277+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.17995v1","created_at":"2026-07-05T08:24:14.307277+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.17995","created_at":"2026-07-05T08:24:14.307277+00:00"},{"alias_kind":"pith_short_12","alias_value":"O2A4P2WD75O4","created_at":"2026-07-05T08:24:14.307277+00:00"},{"alias_kind":"pith_short_16","alias_value":"O2A4P2WD75O4QWFN","created_at":"2026-07-05T08:24:14.307277+00:00"},{"alias_kind":"pith_short_8","alias_value":"O2A4P2WD","created_at":"2026-07-05T08:24:14.307277+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.05586","citing_title":"AeroJEPA: Learning Semantic Latent Representations for Scalable 3D Aerodynamic Field Modeling","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O2A4P2WD75O4QWFN22GMG6ICZQ","json":"https://pith.science/pith/O2A4P2WD75O4QWFN22GMG6ICZQ.json","graph_json":"https://pith.science/api/pith-number/O2A4P2WD75O4QWFN22GMG6ICZQ/graph.json","events_json":"https://pith.science/api/pith-number/O2A4P2WD75O4QWFN22GMG6ICZQ/events.json","paper":"https://pith.science/paper/O2A4P2WD"},"agent_actions":{"view_html":"https://pith.science/pith/O2A4P2WD75O4QWFN22GMG6ICZQ","download_json":"https://pith.science/pith/O2A4P2WD75O4QWFN22GMG6ICZQ.json","view_paper":"https://pith.science/paper/O2A4P2WD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.17995&json=true","fetch_graph":"https://pith.science/api/pith-number/O2A4P2WD75O4QWFN22GMG6ICZQ/graph.json","fetch_events":"https://pith.science/api/pith-number/O2A4P2WD75O4QWFN22GMG6ICZQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O2A4P2WD75O4QWFN22GMG6ICZQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O2A4P2WD75O4QWFN22GMG6ICZQ/action/storage_attestation","attest_author":"https://pith.science/pith/O2A4P2WD75O4QWFN22GMG6ICZQ/action/author_attestation","sign_citation":"https://pith.science/pith/O2A4P2WD75O4QWFN22GMG6ICZQ/action/citation_signature","submit_replication":"https://pith.science/pith/O2A4P2WD75O4QWFN22GMG6ICZQ/action/replication_record"}},"created_at":"2026-07-05T08:24:14.307277+00:00","updated_at":"2026-07-05T08:24:14.307277+00:00"}