{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:2FIF2XGEHK7MNC45JOXGFMQJBI","short_pith_number":"pith:2FIF2XGE","schema_version":"1.0","canonical_sha256":"d1505d5cc43abec68b9d4bae62b2090a351b139ecb3c2bbf5d359b7722f82dce","source":{"kind":"arxiv","id":"2105.11210","version":1},"attestation_state":"computed","paper":{"title":"StructuralLM: Structural Pre-training for Form Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bin Bi, Chenliang Li, Fei Huang, Luo Si, Ming Yan, Songfang Huang, Wei Wang","submitted_at":"2021-05-24T11:33:20Z","abstract_excerpt":"Large pre-trained language models achieve state-of-the-art results when fine-tuned on downstream NLP tasks. However, they almost exclusively focus on text-only representation, while neglecting cell-level layout information that is important for form image understanding. In this paper, we propose a new pre-training approach, StructuralLM, to jointly leverage cell and layout information from scanned documents. Specifically, we pre-train StructuralLM with two new designs to make the most of the interactions of cell and layout information: 1) each cell as a semantic unit; 2) classification of cell"},"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":"2105.11210","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-05-24T11:33:20Z","cross_cats_sorted":[],"title_canon_sha256":"b32b433c08a60bee868bdd5e8537b12b3a6a030477b588c06a12fec90ed4811e","abstract_canon_sha256":"5bc5ae501dc860148ba55eadac6718edfb904a5528a0d5367d0bb2959f6cd120"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:42:36.454758Z","signature_b64":"gSH0Lm2FNPdJ0fc9fvhlIOXZh1uxlhh/HezMcvPOkJzzP7BmB2NzrgluCMZMh9iiizAG/+CKXsSxsmee3G+tAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d1505d5cc43abec68b9d4bae62b2090a351b139ecb3c2bbf5d359b7722f82dce","last_reissued_at":"2026-07-05T02:42:36.454350Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:42:36.454350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"StructuralLM: Structural Pre-training for Form Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bin Bi, Chenliang Li, Fei Huang, Luo Si, Ming Yan, Songfang Huang, Wei Wang","submitted_at":"2021-05-24T11:33:20Z","abstract_excerpt":"Large pre-trained language models achieve state-of-the-art results when fine-tuned on downstream NLP tasks. However, they almost exclusively focus on text-only representation, while neglecting cell-level layout information that is important for form image understanding. In this paper, we propose a new pre-training approach, StructuralLM, to jointly leverage cell and layout information from scanned documents. Specifically, we pre-train StructuralLM with two new designs to make the most of the interactions of cell and layout information: 1) each cell as a semantic unit; 2) classification of cell"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.11210","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/2105.11210/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":"2105.11210","created_at":"2026-07-05T02:42:36.454413+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.11210v1","created_at":"2026-07-05T02:42:36.454413+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.11210","created_at":"2026-07-05T02:42:36.454413+00:00"},{"alias_kind":"pith_short_12","alias_value":"2FIF2XGEHK7M","created_at":"2026-07-05T02:42:36.454413+00:00"},{"alias_kind":"pith_short_16","alias_value":"2FIF2XGEHK7MNC45","created_at":"2026-07-05T02:42:36.454413+00:00"},{"alias_kind":"pith_short_8","alias_value":"2FIF2XGE","created_at":"2026-07-05T02:42:36.454413+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/2FIF2XGEHK7MNC45JOXGFMQJBI","json":"https://pith.science/pith/2FIF2XGEHK7MNC45JOXGFMQJBI.json","graph_json":"https://pith.science/api/pith-number/2FIF2XGEHK7MNC45JOXGFMQJBI/graph.json","events_json":"https://pith.science/api/pith-number/2FIF2XGEHK7MNC45JOXGFMQJBI/events.json","paper":"https://pith.science/paper/2FIF2XGE"},"agent_actions":{"view_html":"https://pith.science/pith/2FIF2XGEHK7MNC45JOXGFMQJBI","download_json":"https://pith.science/pith/2FIF2XGEHK7MNC45JOXGFMQJBI.json","view_paper":"https://pith.science/paper/2FIF2XGE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.11210&json=true","fetch_graph":"https://pith.science/api/pith-number/2FIF2XGEHK7MNC45JOXGFMQJBI/graph.json","fetch_events":"https://pith.science/api/pith-number/2FIF2XGEHK7MNC45JOXGFMQJBI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2FIF2XGEHK7MNC45JOXGFMQJBI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2FIF2XGEHK7MNC45JOXGFMQJBI/action/storage_attestation","attest_author":"https://pith.science/pith/2FIF2XGEHK7MNC45JOXGFMQJBI/action/author_attestation","sign_citation":"https://pith.science/pith/2FIF2XGEHK7MNC45JOXGFMQJBI/action/citation_signature","submit_replication":"https://pith.science/pith/2FIF2XGEHK7MNC45JOXGFMQJBI/action/replication_record"}},"created_at":"2026-07-05T02:42:36.454413+00:00","updated_at":"2026-07-05T02:42:36.454413+00:00"}