{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:EX4WUAYHZGCGLX3A4KXGWH2LFV","short_pith_number":"pith:EX4WUAYH","canonical_record":{"source":{"id":"1912.01774","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-12-04T02:54:18Z","cross_cats_sorted":[],"title_canon_sha256":"aa096c8b0dc90e74d925ecb291385fd77b4f96d662f7131e65eb1e1f9af84e53","abstract_canon_sha256":"feda692486ccd4f930c86fa8905524f32776f84ca17304a7880177aa8debf9f4"},"schema_version":"1.0"},"canonical_sha256":"25f96a0307c98465df60e2ae6b1f4b2d488dbfdbeac445cae39ec058449fad09","source":{"kind":"arxiv","id":"1912.01774","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1912.01774","created_at":"2026-07-05T00:23:53Z"},{"alias_kind":"arxiv_version","alias_value":"1912.01774v1","created_at":"2026-07-05T00:23:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.01774","created_at":"2026-07-05T00:23:53Z"},{"alias_kind":"pith_short_12","alias_value":"EX4WUAYHZGCG","created_at":"2026-07-05T00:23:53Z"},{"alias_kind":"pith_short_16","alias_value":"EX4WUAYHZGCGLX3A","created_at":"2026-07-05T00:23:53Z"},{"alias_kind":"pith_short_8","alias_value":"EX4WUAYH","created_at":"2026-07-05T00:23:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:EX4WUAYHZGCGLX3A4KXGWH2LFV","target":"record","payload":{"canonical_record":{"source":{"id":"1912.01774","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-12-04T02:54:18Z","cross_cats_sorted":[],"title_canon_sha256":"aa096c8b0dc90e74d925ecb291385fd77b4f96d662f7131e65eb1e1f9af84e53","abstract_canon_sha256":"feda692486ccd4f930c86fa8905524f32776f84ca17304a7880177aa8debf9f4"},"schema_version":"1.0"},"canonical_sha256":"25f96a0307c98465df60e2ae6b1f4b2d488dbfdbeac445cae39ec058449fad09","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:23:53.891309Z","signature_b64":"Oj1svf8R+uNL1GOuVDAuFkF4fTpkTP9jNTABNOy8y1QZTbqfnL21o3qH3jQnjcTHs/i15e0UMFymbqrqN/kTDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25f96a0307c98465df60e2ae6b1f4b2d488dbfdbeac445cae39ec058449fad09","last_reissued_at":"2026-07-05T00:23:53.890860Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:23:53.890860Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1912.01774","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T00:23:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MuyTLj/gQNKNx2C5qwkR6TVwiQtoNjK/21NNrDCc1iUn2DnV7517+aJcmCEAzw1IrIy/vZb+0lZY/+IUrdqdCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T09:09:41.901506Z"},"content_sha256":"140199d24cb325e0aa45877c6da73face2e00432bab73222ea3242ce44248052","schema_version":"1.0","event_id":"sha256:140199d24cb325e0aa45877c6da73face2e00432bab73222ea3242ce44248052"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:EX4WUAYHZGCGLX3A4KXGWH2LFV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Acquiring Knowledge from Pre-trained Model to Neural Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Heng Yu, Rongxiang Weng, Shanbo Cheng, Shujian Huang, Weihua Luo","submitted_at":"2019-12-04T02:54:18Z","abstract_excerpt":"Pre-training and fine-tuning have achieved great success in the natural language process field. The standard paradigm of exploiting them includes two steps: first, pre-training a model, e.g. BERT, with a large scale unlabeled monolingual data. Then, fine-tuning the pre-trained model with labeled data from downstream tasks. However, in neural machine translation (NMT), we address the problem that the training objective of the bilingual task is far different from the monolingual pre-trained model. This gap leads that only using fine-tuning in NMT can not fully utilize prior language knowledge. I"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.01774","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/1912.01774/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T00:23:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3ipf18af1y8a3ZGX8pGfEUf9hzwUz5dyp1h4E2VTclATG80pgzmH2c2JtP7agOvlbMoFxY4hLtzVbeOavh2RCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T09:09:41.902407Z"},"content_sha256":"6ac24d5e262df5c71805ffaec5753e7bb9aa5561d66e448a0e4c7a26a32f762e","schema_version":"1.0","event_id":"sha256:6ac24d5e262df5c71805ffaec5753e7bb9aa5561d66e448a0e4c7a26a32f762e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EX4WUAYHZGCGLX3A4KXGWH2LFV/bundle.json","state_url":"https://pith.science/pith/EX4WUAYHZGCGLX3A4KXGWH2LFV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EX4WUAYHZGCGLX3A4KXGWH2LFV/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-05T09:09:41Z","links":{"resolver":"https://pith.science/pith/EX4WUAYHZGCGLX3A4KXGWH2LFV","bundle":"https://pith.science/pith/EX4WUAYHZGCGLX3A4KXGWH2LFV/bundle.json","state":"https://pith.science/pith/EX4WUAYHZGCGLX3A4KXGWH2LFV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EX4WUAYHZGCGLX3A4KXGWH2LFV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:EX4WUAYHZGCGLX3A4KXGWH2LFV","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"feda692486ccd4f930c86fa8905524f32776f84ca17304a7880177aa8debf9f4","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-12-04T02:54:18Z","title_canon_sha256":"aa096c8b0dc90e74d925ecb291385fd77b4f96d662f7131e65eb1e1f9af84e53"},"schema_version":"1.0","source":{"id":"1912.01774","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1912.01774","created_at":"2026-07-05T00:23:53Z"},{"alias_kind":"arxiv_version","alias_value":"1912.01774v1","created_at":"2026-07-05T00:23:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.01774","created_at":"2026-07-05T00:23:53Z"},{"alias_kind":"pith_short_12","alias_value":"EX4WUAYHZGCG","created_at":"2026-07-05T00:23:53Z"},{"alias_kind":"pith_short_16","alias_value":"EX4WUAYHZGCGLX3A","created_at":"2026-07-05T00:23:53Z"},{"alias_kind":"pith_short_8","alias_value":"EX4WUAYH","created_at":"2026-07-05T00:23:53Z"}],"graph_snapshots":[{"event_id":"sha256:6ac24d5e262df5c71805ffaec5753e7bb9aa5561d66e448a0e4c7a26a32f762e","target":"graph","created_at":"2026-07-05T00:23:53Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/1912.01774/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pre-training and fine-tuning have achieved great success in the natural language process field. The standard paradigm of exploiting them includes two steps: first, pre-training a model, e.g. BERT, with a large scale unlabeled monolingual data. Then, fine-tuning the pre-trained model with labeled data from downstream tasks. However, in neural machine translation (NMT), we address the problem that the training objective of the bilingual task is far different from the monolingual pre-trained model. This gap leads that only using fine-tuning in NMT can not fully utilize prior language knowledge. I","authors_text":"Heng Yu, Rongxiang Weng, Shanbo Cheng, Shujian Huang, Weihua Luo","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-12-04T02:54:18Z","title":"Acquiring Knowledge from Pre-trained Model to Neural Machine Translation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.01774","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:140199d24cb325e0aa45877c6da73face2e00432bab73222ea3242ce44248052","target":"record","created_at":"2026-07-05T00:23:53Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"feda692486ccd4f930c86fa8905524f32776f84ca17304a7880177aa8debf9f4","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-12-04T02:54:18Z","title_canon_sha256":"aa096c8b0dc90e74d925ecb291385fd77b4f96d662f7131e65eb1e1f9af84e53"},"schema_version":"1.0","source":{"id":"1912.01774","kind":"arxiv","version":1}},"canonical_sha256":"25f96a0307c98465df60e2ae6b1f4b2d488dbfdbeac445cae39ec058449fad09","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"25f96a0307c98465df60e2ae6b1f4b2d488dbfdbeac445cae39ec058449fad09","first_computed_at":"2026-07-05T00:23:53.890860Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:23:53.890860Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Oj1svf8R+uNL1GOuVDAuFkF4fTpkTP9jNTABNOy8y1QZTbqfnL21o3qH3jQnjcTHs/i15e0UMFymbqrqN/kTDw==","signature_status":"signed_v1","signed_at":"2026-07-05T00:23:53.891309Z","signed_message":"canonical_sha256_bytes"},"source_id":"1912.01774","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:140199d24cb325e0aa45877c6da73face2e00432bab73222ea3242ce44248052","sha256:6ac24d5e262df5c71805ffaec5753e7bb9aa5561d66e448a0e4c7a26a32f762e"],"state_sha256":"ce9c7b7cd2ec0d97e921ea745237e2971d550b7962171ef41b838e9195715bc4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u++Ll1ZKyk+YAV6ZSVsBsTVyL1vq0T4D4cEyia77vcQRLOagwUe5wgseJm6c9vYtsgeMmUOKW8sh04JsJ1eyAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T09:09:41.932659Z","bundle_sha256":"d84ae72dea8abd241714d80da9bbc3b37d9aa3f2b41ff5bc79c24b7df54ca8dc"}}