{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PVSK4TW5XYPJV6K2FDRSUYESMY","short_pith_number":"pith:PVSK4TW5","schema_version":"1.0","canonical_sha256":"7d64ae4eddbe1e9af95a28e32a6092662af9e548d02616458f6b8dbb62ccf34d","source":{"kind":"arxiv","id":"2606.21356","version":1},"attestation_state":"computed","paper":{"title":"SOHET: Sequence Of Heterogeneous Events Transformer with Self-Supervised Pre-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Kees Jan de Vries, Mathijs de Jong, Mustafa Radha","submitted_at":"2026-06-19T11:56:46Z","abstract_excerpt":"Many machine learning applications rely on heterogeneous event streams to make predictions, either causally as events arrive or bidirectionally over complete sequences. We propose SOHET (Sequence Of Heterogeneous Events Transformer), a hierarchical architecture combining event-type-specific tabular encoders with temporal and type embeddings, processed by a causal or bidirectional transformer. We introduce three self-supervised pre-training objectives for the causal setting. On a proprietary large-scale real-world Booking.com fraud detection task with 17 event types, SOHET outperforms FlexTPP, "},"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":"2606.21356","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-06-19T11:56:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"859c35eb488d930cebc8b9d522e1f378f10cb9eda066c3bbd1174c84596121f3","abstract_canon_sha256":"0b3e3aadfa88a390d100c45812afde8db4c0443205f3d7b5817820d5c4767e80"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:12:38.305271Z","signature_b64":"4L8WTj9ursvIklMA0Uqqtrenu2zPDAl/V6H2A5r39Yu9XQV4TlhUp1KHf2JEkbOtds033iq2EmNFAjWEU8MgCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7d64ae4eddbe1e9af95a28e32a6092662af9e548d02616458f6b8dbb62ccf34d","last_reissued_at":"2026-06-23T01:12:38.304815Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:12:38.304815Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SOHET: Sequence Of Heterogeneous Events Transformer with Self-Supervised Pre-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Kees Jan de Vries, Mathijs de Jong, Mustafa Radha","submitted_at":"2026-06-19T11:56:46Z","abstract_excerpt":"Many machine learning applications rely on heterogeneous event streams to make predictions, either causally as events arrive or bidirectionally over complete sequences. We propose SOHET (Sequence Of Heterogeneous Events Transformer), a hierarchical architecture combining event-type-specific tabular encoders with temporal and type embeddings, processed by a causal or bidirectional transformer. We introduce three self-supervised pre-training objectives for the causal setting. On a proprietary large-scale real-world Booking.com fraud detection task with 17 event types, SOHET outperforms FlexTPP, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.21356","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/2606.21356/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":"2606.21356","created_at":"2026-06-23T01:12:38.304888+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.21356v1","created_at":"2026-06-23T01:12:38.304888+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.21356","created_at":"2026-06-23T01:12:38.304888+00:00"},{"alias_kind":"pith_short_12","alias_value":"PVSK4TW5XYPJ","created_at":"2026-06-23T01:12:38.304888+00:00"},{"alias_kind":"pith_short_16","alias_value":"PVSK4TW5XYPJV6K2","created_at":"2026-06-23T01:12:38.304888+00:00"},{"alias_kind":"pith_short_8","alias_value":"PVSK4TW5","created_at":"2026-06-23T01:12:38.304888+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/PVSK4TW5XYPJV6K2FDRSUYESMY","json":"https://pith.science/pith/PVSK4TW5XYPJV6K2FDRSUYESMY.json","graph_json":"https://pith.science/api/pith-number/PVSK4TW5XYPJV6K2FDRSUYESMY/graph.json","events_json":"https://pith.science/api/pith-number/PVSK4TW5XYPJV6K2FDRSUYESMY/events.json","paper":"https://pith.science/paper/PVSK4TW5"},"agent_actions":{"view_html":"https://pith.science/pith/PVSK4TW5XYPJV6K2FDRSUYESMY","download_json":"https://pith.science/pith/PVSK4TW5XYPJV6K2FDRSUYESMY.json","view_paper":"https://pith.science/paper/PVSK4TW5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.21356&json=true","fetch_graph":"https://pith.science/api/pith-number/PVSK4TW5XYPJV6K2FDRSUYESMY/graph.json","fetch_events":"https://pith.science/api/pith-number/PVSK4TW5XYPJV6K2FDRSUYESMY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PVSK4TW5XYPJV6K2FDRSUYESMY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PVSK4TW5XYPJV6K2FDRSUYESMY/action/storage_attestation","attest_author":"https://pith.science/pith/PVSK4TW5XYPJV6K2FDRSUYESMY/action/author_attestation","sign_citation":"https://pith.science/pith/PVSK4TW5XYPJV6K2FDRSUYESMY/action/citation_signature","submit_replication":"https://pith.science/pith/PVSK4TW5XYPJV6K2FDRSUYESMY/action/replication_record"}},"created_at":"2026-06-23T01:12:38.304888+00:00","updated_at":"2026-06-23T01:12:38.304888+00:00"}