{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:ZE3HOWJPNB3CUBIY3CRMZHY3YV","short_pith_number":"pith:ZE3HOWJP","schema_version":"1.0","canonical_sha256":"c93677592f68762a0518d8a2cc9f1bc570e0812bdc32fefeb70349a9f26c53b1","source":{"kind":"arxiv","id":"2203.00281","version":3},"attestation_state":"computed","paper":{"title":"Fast-R2D2: A Pretrained Recursive Neural Network based on Pruned CKY for Grammar Induction and Text Representation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Gerard de Melo, Haitao Mi, Liang Li, Xiang Hu","submitted_at":"2022-03-01T07:54:44Z","abstract_excerpt":"Recently CKY-based models show great potential in unsupervised grammar induction thanks to their human-like encoding paradigm, which runs recursively and hierarchically, but requires $O(n^3)$ time-complexity. Recursive Transformer based on Differentiable Trees (R2D2) makes it possible to scale to large language model pre-training even with complex tree encoder by introducing a heuristic pruning method. However, the rule-based pruning approach suffers from local optimum and slow inference issues. In this paper, we fix those issues in a unified method. We propose to use a top-down parser as a mo"},"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":"2203.00281","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-03-01T07:54:44Z","cross_cats_sorted":[],"title_canon_sha256":"7a39002f62118ec38b6ab889f2472a2c289747919025e5a8d2676c25690d66b5","abstract_canon_sha256":"684f6dfd635a4ce2fe06b18ce30e80c4f40ab44b25ab56b44ef160d2efaa1c13"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:12:27.704292Z","signature_b64":"6/1clUe3iDGnC6OGd7118PstLQn+IJ7N3lka86GdCriLAuwHDCXFQ0HWp2hfG41NO5LjwDNLCcnscxu4mnnjBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c93677592f68762a0518d8a2cc9f1bc570e0812bdc32fefeb70349a9f26c53b1","last_reissued_at":"2026-07-05T05:12:27.703810Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:12:27.703810Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fast-R2D2: A Pretrained Recursive Neural Network based on Pruned CKY for Grammar Induction and Text Representation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Gerard de Melo, Haitao Mi, Liang Li, Xiang Hu","submitted_at":"2022-03-01T07:54:44Z","abstract_excerpt":"Recently CKY-based models show great potential in unsupervised grammar induction thanks to their human-like encoding paradigm, which runs recursively and hierarchically, but requires $O(n^3)$ time-complexity. Recursive Transformer based on Differentiable Trees (R2D2) makes it possible to scale to large language model pre-training even with complex tree encoder by introducing a heuristic pruning method. However, the rule-based pruning approach suffers from local optimum and slow inference issues. In this paper, we fix those issues in a unified method. We propose to use a top-down parser as a mo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.00281","kind":"arxiv","version":3},"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/2203.00281/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":"2203.00281","created_at":"2026-07-05T05:12:27.703871+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.00281v3","created_at":"2026-07-05T05:12:27.703871+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.00281","created_at":"2026-07-05T05:12:27.703871+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZE3HOWJPNB3C","created_at":"2026-07-05T05:12:27.703871+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZE3HOWJPNB3CUBIY","created_at":"2026-07-05T05:12:27.703871+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZE3HOWJP","created_at":"2026-07-05T05:12:27.703871+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/ZE3HOWJPNB3CUBIY3CRMZHY3YV","json":"https://pith.science/pith/ZE3HOWJPNB3CUBIY3CRMZHY3YV.json","graph_json":"https://pith.science/api/pith-number/ZE3HOWJPNB3CUBIY3CRMZHY3YV/graph.json","events_json":"https://pith.science/api/pith-number/ZE3HOWJPNB3CUBIY3CRMZHY3YV/events.json","paper":"https://pith.science/paper/ZE3HOWJP"},"agent_actions":{"view_html":"https://pith.science/pith/ZE3HOWJPNB3CUBIY3CRMZHY3YV","download_json":"https://pith.science/pith/ZE3HOWJPNB3CUBIY3CRMZHY3YV.json","view_paper":"https://pith.science/paper/ZE3HOWJP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.00281&json=true","fetch_graph":"https://pith.science/api/pith-number/ZE3HOWJPNB3CUBIY3CRMZHY3YV/graph.json","fetch_events":"https://pith.science/api/pith-number/ZE3HOWJPNB3CUBIY3CRMZHY3YV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZE3HOWJPNB3CUBIY3CRMZHY3YV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZE3HOWJPNB3CUBIY3CRMZHY3YV/action/storage_attestation","attest_author":"https://pith.science/pith/ZE3HOWJPNB3CUBIY3CRMZHY3YV/action/author_attestation","sign_citation":"https://pith.science/pith/ZE3HOWJPNB3CUBIY3CRMZHY3YV/action/citation_signature","submit_replication":"https://pith.science/pith/ZE3HOWJPNB3CUBIY3CRMZHY3YV/action/replication_record"}},"created_at":"2026-07-05T05:12:27.703871+00:00","updated_at":"2026-07-05T05:12:27.703871+00:00"}