{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:Y27SP73JVV4TMS3KLXHG4XEXPV","short_pith_number":"pith:Y27SP73J","schema_version":"1.0","canonical_sha256":"c6bf27ff69ad79364b6a5dce6e5c977d4aa141b96ca4b26931e60d1c9f297603","source":{"kind":"arxiv","id":"2407.08330","version":1},"attestation_state":"computed","paper":{"title":"HDT: Hierarchical Document Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andreas Geiger, Haoyu He, Iryna Gurevych, Jan Buchmann, Markus Flicke","submitted_at":"2024-07-11T09:28:04Z","abstract_excerpt":"In this paper, we propose the Hierarchical Document Transformer (HDT), a novel sparse Transformer architecture tailored for structured hierarchical documents. Such documents are extremely important in numerous domains, including science, law or medicine. However, most existing solutions are inefficient and fail to make use of the structure inherent to documents. HDT exploits document structure by introducing auxiliary anchor tokens and redesigning the attention mechanism into a sparse multi-level hierarchy. This approach facilitates information exchange between tokens at different levels while"},"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":"2407.08330","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-07-11T09:28:04Z","cross_cats_sorted":[],"title_canon_sha256":"4ac3cbf680b32b7162b16fd4298845c6dcaccda826036f3bc4d86d4ea0faeedf","abstract_canon_sha256":"03ff1a8986a1d0017d58f654dc908ee2fb1380bb8d03335f1f08b1eb75d557ca"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:42:46.844453Z","signature_b64":"zgxwHE4Awo7LoTTy5jGTUiJJM4BrHyVGoGAthpKTX6UxEUOYsLxIF70c5geGYeufWUz/nEekMfMJKUlEtXdtAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c6bf27ff69ad79364b6a5dce6e5c977d4aa141b96ca4b26931e60d1c9f297603","last_reissued_at":"2026-07-05T08:42:46.843993Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:42:46.843993Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HDT: Hierarchical Document Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andreas Geiger, Haoyu He, Iryna Gurevych, Jan Buchmann, Markus Flicke","submitted_at":"2024-07-11T09:28:04Z","abstract_excerpt":"In this paper, we propose the Hierarchical Document Transformer (HDT), a novel sparse Transformer architecture tailored for structured hierarchical documents. Such documents are extremely important in numerous domains, including science, law or medicine. However, most existing solutions are inefficient and fail to make use of the structure inherent to documents. HDT exploits document structure by introducing auxiliary anchor tokens and redesigning the attention mechanism into a sparse multi-level hierarchy. This approach facilitates information exchange between tokens at different levels while"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.08330","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/2407.08330/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":"2407.08330","created_at":"2026-07-05T08:42:46.844052+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.08330v1","created_at":"2026-07-05T08:42:46.844052+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08330","created_at":"2026-07-05T08:42:46.844052+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y27SP73JVV4T","created_at":"2026-07-05T08:42:46.844052+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y27SP73JVV4TMS3K","created_at":"2026-07-05T08:42:46.844052+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y27SP73J","created_at":"2026-07-05T08:42:46.844052+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18525","citing_title":"Overlapping Schwarz Attention: Hierarchical Attention via Domain Decomposition","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2602.00520","citing_title":"NEST: Nested Event Stream Transformer for Sequences of Multisets","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y27SP73JVV4TMS3KLXHG4XEXPV","json":"https://pith.science/pith/Y27SP73JVV4TMS3KLXHG4XEXPV.json","graph_json":"https://pith.science/api/pith-number/Y27SP73JVV4TMS3KLXHG4XEXPV/graph.json","events_json":"https://pith.science/api/pith-number/Y27SP73JVV4TMS3KLXHG4XEXPV/events.json","paper":"https://pith.science/paper/Y27SP73J"},"agent_actions":{"view_html":"https://pith.science/pith/Y27SP73JVV4TMS3KLXHG4XEXPV","download_json":"https://pith.science/pith/Y27SP73JVV4TMS3KLXHG4XEXPV.json","view_paper":"https://pith.science/paper/Y27SP73J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.08330&json=true","fetch_graph":"https://pith.science/api/pith-number/Y27SP73JVV4TMS3KLXHG4XEXPV/graph.json","fetch_events":"https://pith.science/api/pith-number/Y27SP73JVV4TMS3KLXHG4XEXPV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y27SP73JVV4TMS3KLXHG4XEXPV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y27SP73JVV4TMS3KLXHG4XEXPV/action/storage_attestation","attest_author":"https://pith.science/pith/Y27SP73JVV4TMS3KLXHG4XEXPV/action/author_attestation","sign_citation":"https://pith.science/pith/Y27SP73JVV4TMS3KLXHG4XEXPV/action/citation_signature","submit_replication":"https://pith.science/pith/Y27SP73JVV4TMS3KLXHG4XEXPV/action/replication_record"}},"created_at":"2026-07-05T08:42:46.844052+00:00","updated_at":"2026-07-05T08:42:46.844052+00:00"}