{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:BAORJIUMBMALGSLINQYG3O36Q6","short_pith_number":"pith:BAORJIUM","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"},"canonical_sha256":"081d14a28c0b00b349686c306dbb7e87ab73e8cf14ae61c95988df9fdadf341c","source":{"kind":"arxiv","id":"2101.08106","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2101.08106","created_at":"2026-07-05T02:50:43Z"},{"alias_kind":"arxiv_version","alias_value":"2101.08106v2","created_at":"2026-07-05T02:50:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.08106","created_at":"2026-07-05T02:50:43Z"},{"alias_kind":"pith_short_12","alias_value":"BAORJIUMBMAL","created_at":"2026-07-05T02:50:43Z"},{"alias_kind":"pith_short_16","alias_value":"BAORJIUMBMALGSLI","created_at":"2026-07-05T02:50:43Z"},{"alias_kind":"pith_short_8","alias_value":"BAORJIUM","created_at":"2026-07-05T02:50:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:BAORJIUMBMALGSLINQYG3O36Q6","target":"record","payload":{"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"},"canonical_sha256":"081d14a28c0b00b349686c306dbb7e87ab73e8cf14ae61c95988df9fdadf341c","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"},"source_kind":"arxiv","source_id":"2101.08106","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:50:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ITBcEKP2nSOKOfthVD60mfjMgTKZl8uijnv42Cf+CLcOMGrQ3wxg9NaLF4iYDPuVdS2ZN1mVBZOo4m3NQNPhBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T14:41:27.212338Z"},"content_sha256":"0540ff2a3cf0406d70e04f45f3fdba838be432d4af11a2da3f33506a9053bc71","schema_version":"1.0","event_id":"sha256:0540ff2a3cf0406d70e04f45f3fdba838be432d4af11a2da3f33506a9053bc71"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:BAORJIUMBMALGSLINQYG3O36Q6","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:50:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0taLWNAEMJtYiTEMCiyrlxAEjoRR/QUlYGjwhJJPFBqqLKrkl7y3JkOVFdkJMEuuzQu9MQm1IwZBev2dgRy/Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T14:41:27.212703Z"},"content_sha256":"c2a4095bb66032a6fdb695cf81ff370089369bd7d0d5105a38cd6fc260bcc320","schema_version":"1.0","event_id":"sha256:c2a4095bb66032a6fdb695cf81ff370089369bd7d0d5105a38cd6fc260bcc320"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6/bundle.json","state_url":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BAORJIUMBMALGSLINQYG3O36Q6/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":"2026-07-07T14:41:27Z","links":{"resolver":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6","bundle":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6/bundle.json","state":"https://pith.science/pith/BAORJIUMBMALGSLINQYG3O36Q6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BAORJIUMBMALGSLINQYG3O36Q6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:BAORJIUMBMALGSLINQYG3O36Q6","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":"4b6b85fcad285f10efaa104e95a111814e938825b98f8ede5fa1d84934b242d8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-01-20T13:07:39Z","title_canon_sha256":"896d675316e798696789d0b4c4f0f236a7340ab27ac9df0ea5a20c8e2246c151"},"schema_version":"1.0","source":{"id":"2101.08106","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2101.08106","created_at":"2026-07-05T02:50:43Z"},{"alias_kind":"arxiv_version","alias_value":"2101.08106v2","created_at":"2026-07-05T02:50:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.08106","created_at":"2026-07-05T02:50:43Z"},{"alias_kind":"pith_short_12","alias_value":"BAORJIUMBMAL","created_at":"2026-07-05T02:50:43Z"},{"alias_kind":"pith_short_16","alias_value":"BAORJIUMBMALGSLI","created_at":"2026-07-05T02:50:43Z"},{"alias_kind":"pith_short_8","alias_value":"BAORJIUM","created_at":"2026-07-05T02:50:43Z"}],"graph_snapshots":[{"event_id":"sha256:c2a4095bb66032a6fdb695cf81ff370089369bd7d0d5105a38cd6fc260bcc320","target":"graph","created_at":"2026-07-05T02:50:43Z","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/2101.08106/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Hai-Tao Zheng, Lingyun Feng, Minghui Qiu, Yaliang Li, Ying Shen","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-01-20T13:07:39Z","title":"Learning to Augment for Data-Scarce Domain BERT Knowledge Distillation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2101.08106","kind":"arxiv","version":2},"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":"sha256:0540ff2a3cf0406d70e04f45f3fdba838be432d4af11a2da3f33506a9053bc71","target":"record","created_at":"2026-07-05T02:50:43Z","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":"4b6b85fcad285f10efaa104e95a111814e938825b98f8ede5fa1d84934b242d8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-01-20T13:07:39Z","title_canon_sha256":"896d675316e798696789d0b4c4f0f236a7340ab27ac9df0ea5a20c8e2246c151"},"schema_version":"1.0","source":{"id":"2101.08106","kind":"arxiv","version":2}},"canonical_sha256":"081d14a28c0b00b349686c306dbb7e87ab73e8cf14ae61c95988df9fdadf341c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"081d14a28c0b00b349686c306dbb7e87ab73e8cf14ae61c95988df9fdadf341c","first_computed_at":"2026-07-05T02:50:43.172817Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:50:43.172817Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7E3O9p4K+s/XR2KP6GSkhg0BoDQH70ch6oFAy4JS8pE4ALq4J58Hc8lrsol7osDMhRXCW1EP9mFK88d+3f98Aw==","signature_status":"signed_v1","signed_at":"2026-07-05T02:50:43.173341Z","signed_message":"canonical_sha256_bytes"},"source_id":"2101.08106","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0540ff2a3cf0406d70e04f45f3fdba838be432d4af11a2da3f33506a9053bc71","sha256:c2a4095bb66032a6fdb695cf81ff370089369bd7d0d5105a38cd6fc260bcc320"],"state_sha256":"51fed7e5e3bc889ce5f7e757456d0300bfc51398e5ed5fa847f966000932984b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oLoMfMJgRiXFVbl9BPFtwmK850CyfPz3bofSCHy2K1ZxL3LfVyfD3NQykF+L7cA5QUM3EoliQZoZvpC8/19hCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T14:41:27.215047Z","bundle_sha256":"d20589e0e33d5a22ced1c6e7c974036e0e2470963a3c9c92d30d05730def1c9c"}}