{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:DTED5ZYBWAL73BBPJGLBUOOYMB","short_pith_number":"pith:DTED5ZYB","schema_version":"1.0","canonical_sha256":"1cc83ee701b017fd842f49961a39d86040e14fda34e55e322f0d6ac0c83716b8","source":{"kind":"arxiv","id":"2106.00479","version":1},"attestation_state":"computed","paper":{"title":"DoT: An efficient Double Transformer for NLP tasks with tables","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Julian Martin Eisenschlos, Syrine Krichene, Thomas M\\\"uller","submitted_at":"2021-06-01T13:33:53Z","abstract_excerpt":"Transformer-based approaches have been successfully used to obtain state-of-the-art accuracy on natural language processing (NLP) tasks with semi-structured tables. These model architectures are typically deep, resulting in slow training and inference, especially for long inputs. To improve efficiency while maintaining a high accuracy, we propose a new architecture, DoT, a double transformer model, that decomposes the problem into two sub-tasks: A shallow pruning transformer that selects the top-K tokens, followed by a deep task-specific transformer that takes as input those K tokens. Addition"},"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":"2106.00479","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-06-01T13:33:53Z","cross_cats_sorted":[],"title_canon_sha256":"4c145b4b10a176042ac1d31d72fbee6b0e5c63656845aa90663648bbd8059ca0","abstract_canon_sha256":"c96c6b719568caad53bf3f3781bd096a8e45110932d46ac073ca1a955649d776"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:45:05.046190Z","signature_b64":"GL7Zgjy7Zc7X0ZS0cmGdfMjCuxdKgTDMMw7SYar6EcgnltVIVP1perlybiBTZwjlAP815rYVJFRq57twpbDcDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1cc83ee701b017fd842f49961a39d86040e14fda34e55e322f0d6ac0c83716b8","last_reissued_at":"2026-07-05T02:45:05.045748Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:45:05.045748Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DoT: An efficient Double Transformer for NLP tasks with tables","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Julian Martin Eisenschlos, Syrine Krichene, Thomas M\\\"uller","submitted_at":"2021-06-01T13:33:53Z","abstract_excerpt":"Transformer-based approaches have been successfully used to obtain state-of-the-art accuracy on natural language processing (NLP) tasks with semi-structured tables. These model architectures are typically deep, resulting in slow training and inference, especially for long inputs. To improve efficiency while maintaining a high accuracy, we propose a new architecture, DoT, a double transformer model, that decomposes the problem into two sub-tasks: A shallow pruning transformer that selects the top-K tokens, followed by a deep task-specific transformer that takes as input those K tokens. Addition"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.00479","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/2106.00479/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":"2106.00479","created_at":"2026-07-05T02:45:05.045815+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.00479v1","created_at":"2026-07-05T02:45:05.045815+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.00479","created_at":"2026-07-05T02:45:05.045815+00:00"},{"alias_kind":"pith_short_12","alias_value":"DTED5ZYBWAL7","created_at":"2026-07-05T02:45:05.045815+00:00"},{"alias_kind":"pith_short_16","alias_value":"DTED5ZYBWAL73BBP","created_at":"2026-07-05T02:45:05.045815+00:00"},{"alias_kind":"pith_short_8","alias_value":"DTED5ZYB","created_at":"2026-07-05T02:45:05.045815+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/DTED5ZYBWAL73BBPJGLBUOOYMB","json":"https://pith.science/pith/DTED5ZYBWAL73BBPJGLBUOOYMB.json","graph_json":"https://pith.science/api/pith-number/DTED5ZYBWAL73BBPJGLBUOOYMB/graph.json","events_json":"https://pith.science/api/pith-number/DTED5ZYBWAL73BBPJGLBUOOYMB/events.json","paper":"https://pith.science/paper/DTED5ZYB"},"agent_actions":{"view_html":"https://pith.science/pith/DTED5ZYBWAL73BBPJGLBUOOYMB","download_json":"https://pith.science/pith/DTED5ZYBWAL73BBPJGLBUOOYMB.json","view_paper":"https://pith.science/paper/DTED5ZYB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.00479&json=true","fetch_graph":"https://pith.science/api/pith-number/DTED5ZYBWAL73BBPJGLBUOOYMB/graph.json","fetch_events":"https://pith.science/api/pith-number/DTED5ZYBWAL73BBPJGLBUOOYMB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DTED5ZYBWAL73BBPJGLBUOOYMB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DTED5ZYBWAL73BBPJGLBUOOYMB/action/storage_attestation","attest_author":"https://pith.science/pith/DTED5ZYBWAL73BBPJGLBUOOYMB/action/author_attestation","sign_citation":"https://pith.science/pith/DTED5ZYBWAL73BBPJGLBUOOYMB/action/citation_signature","submit_replication":"https://pith.science/pith/DTED5ZYBWAL73BBPJGLBUOOYMB/action/replication_record"}},"created_at":"2026-07-05T02:45:05.045815+00:00","updated_at":"2026-07-05T02:45:05.045815+00:00"}