{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:BAORJIUMBMALGSLINQYG3O36Q6","short_pith_number":"pith:BAORJIUM","schema_version":"1.0","canonical_sha256":"081d14a28c0b00b349686c306dbb7e87ab73e8cf14ae61c95988df9fdadf341c","source":{"kind":"arxiv","id":"2101.08106","version":2},"attestation_state":"computed","paper":{"title":"Learning to Augment for Data-Scarce Domain BERT Knowledge Distillation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hai-Tao Zheng, Lingyun Feng, Minghui Qiu, Yaliang Li, Ying Shen","submitted_at":"2021-01-20T13:07:39Z","abstract_excerpt":"Despite pre-trained language models such as BERT have achieved appealing performance in a wide range of natural language processing tasks, they are computationally expensive to be deployed in real-time applications. A typical method is to adopt knowledge distillation to compress these large pre-trained models (teacher models) to small student models. However, for a target domain with scarce training data, the teacher can hardly pass useful knowledge to the student, which yields performance degradation for the student models. To tackle this problem, we propose a method to learn to augment for d"},"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":"2101.08106","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-01-20T13:07:39Z","cross_cats_sorted":[],"title_canon_sha256":"896d675316e798696789d0b4c4f0f236a7340ab27ac9df0ea5a20c8e2246c151","abstract_canon_sha256":"4b6b85fcad285f10efaa104e95a111814e938825b98f8ede5fa1d84934b242d8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:50:43.173341Z","signature_b64":"7E3O9p4K+s/XR2KP6GSkhg0BoDQH70ch6oFAy4JS8pE4ALq4J58Hc8lrsol7osDMhRXCW1EP9mFK88d+3f98Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"081d14a28c0b00b349686c306dbb7e87ab73e8cf14ae61c95988df9fdadf341c","last_reissued_at":"2026-07-05T02:50:43.172817Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:50:43.172817Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Augment for Data-Scarce Domain BERT Knowledge Distillation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hai-Tao Zheng, Lingyun Feng, Minghui Qiu, Yaliang Li, Ying Shen","submitted_at":"2021-01-20T13:07:39Z","abstract_excerpt":"Despite pre-trained language models such as BERT have achieved appealing performance in a wide range of natural language processing tasks, they are computationally expensive to be deployed in real-time applications. A typical method is to adopt knowledge distillation to compress these large pre-trained models (teacher models) to small student models. However, for a target domain with scarce training data, the teacher can hardly pass useful knowledge to the student, which yields performance degradation for the student models. To tackle this problem, we propose a method to learn to augment for d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2101.08106","kind":"arxiv","version":2},"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/2101.08106/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":"2101.08106","created_at":"2026-07-05T02:50:43.172872+00:00"},{"alias_kind":"arxiv_version","alias_value":"2101.08106v2","created_at":"2026-07-05T02:50:43.172872+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.08106","created_at":"2026-07-05T02:50:43.172872+00:00"},{"alias_kind":"pith_short_12","alias_value":"BAORJIUMBMAL","created_at":"2026-07-05T02:50:43.172872+00:00"},{"alias_kind":"pith_short_16","alias_value":"BAORJIUMBMALGSLI","created_at":"2026-07-05T02:50:43.172872+00:00"},{"alias_kind":"pith_short_8","alias_value":"BAORJIUM","created_at":"2026-07-05T02:50:43.172872+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/BAORJIUMBMALGSLINQYG3O36Q6","json":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6.json","graph_json":"https://pith.science/api/pith-number/BAORJIUMBMALGSLINQYG3O36Q6/graph.json","events_json":"https://pith.science/api/pith-number/BAORJIUMBMALGSLINQYG3O36Q6/events.json","paper":"https://pith.science/paper/BAORJIUM"},"agent_actions":{"view_html":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6","download_json":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6.json","view_paper":"https://pith.science/paper/BAORJIUM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2101.08106&json=true","fetch_graph":"https://pith.science/api/pith-number/BAORJIUMBMALGSLINQYG3O36Q6/graph.json","fetch_events":"https://pith.science/api/pith-number/BAORJIUMBMALGSLINQYG3O36Q6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6/action/storage_attestation","attest_author":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6/action/author_attestation","sign_citation":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6/action/citation_signature","submit_replication":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6/action/replication_record"}},"created_at":"2026-07-05T02:50:43.172872+00:00","updated_at":"2026-07-05T02:50:43.172872+00:00"}