{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:67I7MYR5IQZZSUMPGSLSBWIBHN","short_pith_number":"pith:67I7MYR5","canonical_record":{"source":{"id":"1908.09355","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-08-25T16:13:24Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"cdf876163c7f0d041b4058d62bbfcc6e1b78281bd7b312a4dcb5de9087d1daa8","abstract_canon_sha256":"6e7b1e6881aca8b0dd1153aeffe379964b8452343e882fe340abe1622a29a048"},"schema_version":"1.0"},"canonical_sha256":"f7d1f6623d443399518f349720d9013b567711e113fc99b564afb207fd072af1","source":{"kind":"arxiv","id":"1908.09355","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1908.09355","created_at":"2026-07-04T23:59:39Z"},{"alias_kind":"arxiv_version","alias_value":"1908.09355v1","created_at":"2026-07-04T23:59:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.09355","created_at":"2026-07-04T23:59:39Z"},{"alias_kind":"pith_short_12","alias_value":"67I7MYR5IQZZ","created_at":"2026-07-04T23:59:39Z"},{"alias_kind":"pith_short_16","alias_value":"67I7MYR5IQZZSUMP","created_at":"2026-07-04T23:59:39Z"},{"alias_kind":"pith_short_8","alias_value":"67I7MYR5","created_at":"2026-07-04T23:59:39Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:67I7MYR5IQZZSUMPGSLSBWIBHN","target":"record","payload":{"canonical_record":{"source":{"id":"1908.09355","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-08-25T16:13:24Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"cdf876163c7f0d041b4058d62bbfcc6e1b78281bd7b312a4dcb5de9087d1daa8","abstract_canon_sha256":"6e7b1e6881aca8b0dd1153aeffe379964b8452343e882fe340abe1622a29a048"},"schema_version":"1.0"},"canonical_sha256":"f7d1f6623d443399518f349720d9013b567711e113fc99b564afb207fd072af1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:59:39.416641Z","signature_b64":"O4mFqya+L6nu5ck7AnoOOd1InwSubgmPwP7bFrizm6HinC+RlvvOHEVfPbVO/qCG0MtOr2NwKzQwzsOFTXnLBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f7d1f6623d443399518f349720d9013b567711e113fc99b564afb207fd072af1","last_reissued_at":"2026-07-04T23:59:39.416163Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:59:39.416163Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1908.09355","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":"2026-07-04T23:59:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BuwpadhU7Hy6tvIcjQVCMDUIrvh4wE4JepxkTH6gG1n8VjwvLFTFblW0rcrpTUqleVDsJICYo76hnFi9ykkTCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T19:21:49.278641Z"},"content_sha256":"f2a3f0571c8d398a742d623ef4526e5605b789e59a383d1786aad1a7f0794e8c","schema_version":"1.0","event_id":"sha256:f2a3f0571c8d398a742d623ef4526e5605b789e59a383d1786aad1a7f0794e8c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:67I7MYR5IQZZSUMPGSLSBWIBHN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Patient Knowledge Distillation for BERT Model Compression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Jingjing Liu, Siqi Sun, Yu Cheng, Zhe Gan","submitted_at":"2019-08-25T16:13:24Z","abstract_excerpt":"Pre-trained language models such as BERT have proven to be highly effective for natural language processing (NLP) tasks. However, the high demand for computing resources in training such models hinders their application in practice. In order to alleviate this resource hunger in large-scale model training, we propose a Patient Knowledge Distillation approach to compress an original large model (teacher) into an equally-effective lightweight shallow network (student). Different from previous knowledge distillation methods, which only use the output from the last layer of the teacher network for "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.09355","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/1908.09355/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":"2026-07-04T23:59:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yqYkDOyFs6zM6HF1DugHnSELHPR4nNxhrfHzm+/J8KUqPAYJKr/y0l/8omry3/rFT+CapgMA8ca1lrOZqc9zDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T19:21:49.279370Z"},"content_sha256":"b4cfd4f97014fc89c84a1ce5d3d833ffe5a7cc8a5e5449de7d50688cfca244ab","schema_version":"1.0","event_id":"sha256:b4cfd4f97014fc89c84a1ce5d3d833ffe5a7cc8a5e5449de7d50688cfca244ab"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/67I7MYR5IQZZSUMPGSLSBWIBHN/bundle.json","state_url":"https://pith.science/pith/67I7MYR5IQZZSUMPGSLSBWIBHN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/67I7MYR5IQZZSUMPGSLSBWIBHN/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":"2026-08-03T19:21:49Z","links":{"resolver":"https://pith.science/pith/67I7MYR5IQZZSUMPGSLSBWIBHN","bundle":"https://pith.science/pith/67I7MYR5IQZZSUMPGSLSBWIBHN/bundle.json","state":"https://pith.science/pith/67I7MYR5IQZZSUMPGSLSBWIBHN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/67I7MYR5IQZZSUMPGSLSBWIBHN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:67I7MYR5IQZZSUMPGSLSBWIBHN","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":"6e7b1e6881aca8b0dd1153aeffe379964b8452343e882fe340abe1622a29a048","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-08-25T16:13:24Z","title_canon_sha256":"cdf876163c7f0d041b4058d62bbfcc6e1b78281bd7b312a4dcb5de9087d1daa8"},"schema_version":"1.0","source":{"id":"1908.09355","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1908.09355","created_at":"2026-07-04T23:59:39Z"},{"alias_kind":"arxiv_version","alias_value":"1908.09355v1","created_at":"2026-07-04T23:59:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.09355","created_at":"2026-07-04T23:59:39Z"},{"alias_kind":"pith_short_12","alias_value":"67I7MYR5IQZZ","created_at":"2026-07-04T23:59:39Z"},{"alias_kind":"pith_short_16","alias_value":"67I7MYR5IQZZSUMP","created_at":"2026-07-04T23:59:39Z"},{"alias_kind":"pith_short_8","alias_value":"67I7MYR5","created_at":"2026-07-04T23:59:39Z"}],"graph_snapshots":[{"event_id":"sha256:b4cfd4f97014fc89c84a1ce5d3d833ffe5a7cc8a5e5449de7d50688cfca244ab","target":"graph","created_at":"2026-07-04T23:59:39Z","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/1908.09355/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pre-trained language models such as BERT have proven to be highly effective for natural language processing (NLP) tasks. However, the high demand for computing resources in training such models hinders their application in practice. In order to alleviate this resource hunger in large-scale model training, we propose a Patient Knowledge Distillation approach to compress an original large model (teacher) into an equally-effective lightweight shallow network (student). Different from previous knowledge distillation methods, which only use the output from the last layer of the teacher network for ","authors_text":"Jingjing Liu, Siqi Sun, Yu Cheng, Zhe Gan","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-08-25T16:13:24Z","title":"Patient Knowledge Distillation for BERT Model Compression"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.09355","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:f2a3f0571c8d398a742d623ef4526e5605b789e59a383d1786aad1a7f0794e8c","target":"record","created_at":"2026-07-04T23:59:39Z","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":"6e7b1e6881aca8b0dd1153aeffe379964b8452343e882fe340abe1622a29a048","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-08-25T16:13:24Z","title_canon_sha256":"cdf876163c7f0d041b4058d62bbfcc6e1b78281bd7b312a4dcb5de9087d1daa8"},"schema_version":"1.0","source":{"id":"1908.09355","kind":"arxiv","version":1}},"canonical_sha256":"f7d1f6623d443399518f349720d9013b567711e113fc99b564afb207fd072af1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f7d1f6623d443399518f349720d9013b567711e113fc99b564afb207fd072af1","first_computed_at":"2026-07-04T23:59:39.416163Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-04T23:59:39.416163Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"O4mFqya+L6nu5ck7AnoOOd1InwSubgmPwP7bFrizm6HinC+RlvvOHEVfPbVO/qCG0MtOr2NwKzQwzsOFTXnLBA==","signature_status":"signed_v1","signed_at":"2026-07-04T23:59:39.416641Z","signed_message":"canonical_sha256_bytes"},"source_id":"1908.09355","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f2a3f0571c8d398a742d623ef4526e5605b789e59a383d1786aad1a7f0794e8c","sha256:b4cfd4f97014fc89c84a1ce5d3d833ffe5a7cc8a5e5449de7d50688cfca244ab"],"state_sha256":"9eff773b48ee1e33d736fe0809782ea9359b077d33937f2e1cb1930e0b3537a2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Tg6w8AUpv6TJCIsEQAKFv4j6UEyTHQnPB48XcaWEokcYEDkDTXj+GqkSB7zZNCF2OfutIMVED6q38B/sFcSMDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T19:21:49.285552Z","bundle_sha256":"b0352c911b5d5535a2283e09204a332f4980be6b3ff281c6167ee014a6a70773"}}