{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CMJUQXF64I5MLESWSHW2QAIHXQ","short_pith_number":"pith:CMJUQXF6","schema_version":"1.0","canonical_sha256":"1313485cbee23ac5925691eda80107bc1a3ec31cf7d04ddda6d7710c0ec65471","source":{"kind":"arxiv","id":"2605.21490","version":1},"attestation_state":"computed","paper":{"title":"Temporal Contrastive Transformer for Financial Crime Detection: Self-Supervised Sequence Embeddings via Predictive Contrastive Coding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Danny Butvinik (NICE Actimize), Gabrielle Azoulay (NICE Actimize), Nitzan Tal (NICE Actimize), Yonit Marcus (NICE Actimize)","submitted_at":"2026-03-31T09:42:08Z","abstract_excerpt":"We introduce the Temporal Contrastive Transformer (TCT), a representation learning framework designed to capture contextual temporal dynamics in sequences of financial transactions. The model is trained using a self-supervised contrastive objective to produce embeddings that encode behavioral patterns over time, with the goal of supporting downstream fraud detection tasks. We evaluate TCT in a realistic setting by using the learned embeddings as input features to a gradient boosting classifier. Experimental results show that embeddings alone achieve meaningful predictive performance (AUC 0.864"},"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":"2605.21490","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-03-31T09:42:08Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"db5f9c6ad2060f0a737dca149efb8a8cb1517ec91e7f1f42c150a1233a0c9899","abstract_canon_sha256":"2b8b1f7566b115f0f8f31262344302812ffc620c9930b633ebf8691a7b90ae5e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T00:02:26.782729Z","signature_b64":"DBtCbHDWT4VUwqiHuIBSSTlY6FJvLpUMD9qLdtKx7MtPozvzY5Byg7J5U7Hl5teqUWfYHg0qh7+PTgMdANIACw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1313485cbee23ac5925691eda80107bc1a3ec31cf7d04ddda6d7710c0ec65471","last_reissued_at":"2026-05-22T00:02:26.782179Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T00:02:26.782179Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Temporal Contrastive Transformer for Financial Crime Detection: Self-Supervised Sequence Embeddings via Predictive Contrastive Coding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Danny Butvinik (NICE Actimize), Gabrielle Azoulay (NICE Actimize), Nitzan Tal (NICE Actimize), Yonit Marcus (NICE Actimize)","submitted_at":"2026-03-31T09:42:08Z","abstract_excerpt":"We introduce the Temporal Contrastive Transformer (TCT), a representation learning framework designed to capture contextual temporal dynamics in sequences of financial transactions. The model is trained using a self-supervised contrastive objective to produce embeddings that encode behavioral patterns over time, with the goal of supporting downstream fraud detection tasks. We evaluate TCT in a realistic setting by using the learned embeddings as input features to a gradient boosting classifier. Experimental results show that embeddings alone achieve meaningful predictive performance (AUC 0.864"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21490","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/2605.21490/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":"2605.21490","created_at":"2026-05-22T00:02:26.782248+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.21490v1","created_at":"2026-05-22T00:02:26.782248+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21490","created_at":"2026-05-22T00:02:26.782248+00:00"},{"alias_kind":"pith_short_12","alias_value":"CMJUQXF64I5M","created_at":"2026-05-22T00:02:26.782248+00:00"},{"alias_kind":"pith_short_16","alias_value":"CMJUQXF64I5MLESW","created_at":"2026-05-22T00:02:26.782248+00:00"},{"alias_kind":"pith_short_8","alias_value":"CMJUQXF6","created_at":"2026-05-22T00:02:26.782248+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CMJUQXF64I5MLESWSHW2QAIHXQ","json":"https://pith.science/pith/CMJUQXF64I5MLESWSHW2QAIHXQ.json","graph_json":"https://pith.science/api/pith-number/CMJUQXF64I5MLESWSHW2QAIHXQ/graph.json","events_json":"https://pith.science/api/pith-number/CMJUQXF64I5MLESWSHW2QAIHXQ/events.json","paper":"https://pith.science/paper/CMJUQXF6"},"agent_actions":{"view_html":"https://pith.science/pith/CMJUQXF64I5MLESWSHW2QAIHXQ","download_json":"https://pith.science/pith/CMJUQXF64I5MLESWSHW2QAIHXQ.json","view_paper":"https://pith.science/paper/CMJUQXF6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.21490&json=true","fetch_graph":"https://pith.science/api/pith-number/CMJUQXF64I5MLESWSHW2QAIHXQ/graph.json","fetch_events":"https://pith.science/api/pith-number/CMJUQXF64I5MLESWSHW2QAIHXQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CMJUQXF64I5MLESWSHW2QAIHXQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CMJUQXF64I5MLESWSHW2QAIHXQ/action/storage_attestation","attest_author":"https://pith.science/pith/CMJUQXF64I5MLESWSHW2QAIHXQ/action/author_attestation","sign_citation":"https://pith.science/pith/CMJUQXF64I5MLESWSHW2QAIHXQ/action/citation_signature","submit_replication":"https://pith.science/pith/CMJUQXF64I5MLESWSHW2QAIHXQ/action/replication_record"}},"created_at":"2026-05-22T00:02:26.782248+00:00","updated_at":"2026-05-22T00:02:26.782248+00:00"}