{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:TTMP2ELCNS2NE3U7BOG3JLLU3U","short_pith_number":"pith:TTMP2ELC","canonical_record":{"source":{"id":"2106.01023","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-02T08:42:33Z","cross_cats_sorted":[],"title_canon_sha256":"08991a327db560e5a6db6adaf624791eb74809855484576470684b7a622370c2","abstract_canon_sha256":"f4d84ce92e1adb54d3a9096a8d8106827bffb04d4949d4bdf4c395d1fe418287"},"schema_version":"1.0"},"canonical_sha256":"9cd8fd11626cb4d26e9f0b8db4ad74dd27a9d08c19da3cf020f6048edaa3ab9f","source":{"kind":"arxiv","id":"2106.01023","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.01023","created_at":"2026-07-05T02:45:42Z"},{"alias_kind":"arxiv_version","alias_value":"2106.01023v1","created_at":"2026-07-05T02:45:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.01023","created_at":"2026-07-05T02:45:42Z"},{"alias_kind":"pith_short_12","alias_value":"TTMP2ELCNS2N","created_at":"2026-07-05T02:45:42Z"},{"alias_kind":"pith_short_16","alias_value":"TTMP2ELCNS2NE3U7","created_at":"2026-07-05T02:45:42Z"},{"alias_kind":"pith_short_8","alias_value":"TTMP2ELC","created_at":"2026-07-05T02:45:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:TTMP2ELCNS2NE3U7BOG3JLLU3U","target":"record","payload":{"canonical_record":{"source":{"id":"2106.01023","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-02T08:42:33Z","cross_cats_sorted":[],"title_canon_sha256":"08991a327db560e5a6db6adaf624791eb74809855484576470684b7a622370c2","abstract_canon_sha256":"f4d84ce92e1adb54d3a9096a8d8106827bffb04d4949d4bdf4c395d1fe418287"},"schema_version":"1.0"},"canonical_sha256":"9cd8fd11626cb4d26e9f0b8db4ad74dd27a9d08c19da3cf020f6048edaa3ab9f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:45:42.980039Z","signature_b64":"iV7/b90JucmE+fnnx9p5Rjq4aW8fOcx1Xp03goNslMh3eQfnMp9Rhw4/QC/CVa8VqtyJimoWL+3djnEtmesFDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9cd8fd11626cb4d26e9f0b8db4ad74dd27a9d08c19da3cf020f6048edaa3ab9f","last_reissued_at":"2026-07-05T02:45:42.979614Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:45:42.979614Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2106.01023","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":"2026-07-05T02:45:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cN6WRzxjG5QbAJvliJtOgpSdoUM42CpFw+/GXi4y5eHuuGnGKXRuIt8Ls9g051hcAffn97eoZtFsS2uTJnCeCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T19:56:21.835773Z"},"content_sha256":"ef8a15e3d71722d61210c1b2d828f4e80b38ace5346b693ab1399f50fd75395d","schema_version":"1.0","event_id":"sha256:ef8a15e3d71722d61210c1b2d828f4e80b38ace5346b693ab1399f50fd75395d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:TTMP2ELCNS2NE3U7BOG3JLLU3U","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"One Teacher is Enough? Pre-trained Language Model Distillation from Multiple Teachers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chuhan Wu, Fangzhao Wu, Yongfeng Huang","submitted_at":"2021-06-02T08:42:33Z","abstract_excerpt":"Pre-trained language models (PLMs) achieve great success in NLP. However, their huge model sizes hinder their applications in many practical systems. Knowledge distillation is a popular technique to compress PLMs, which learns a small student model from a large teacher PLM. However, the knowledge learned from a single teacher may be limited and even biased, resulting in low-quality student model. In this paper, we propose a multi-teacher knowledge distillation framework named MT-BERT for pre-trained language model compression, which can train high-quality student model from multiple teacher PL"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.01023","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/2106.01023/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":"2026-07-05T02:45:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/m5DXhrtJbeOmUL27pzBu0mXpYU0GQpVGoBgS23J6YNCGb3zaH65WbmVpJ1Gf/s1LVL8IVJrX3WPyc1VBWQYAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T19:56:21.836348Z"},"content_sha256":"dcc673aedba14ded79a9525b23673da8c0f43c23285e12b32ebb0cf9bf8ae782","schema_version":"1.0","event_id":"sha256:dcc673aedba14ded79a9525b23673da8c0f43c23285e12b32ebb0cf9bf8ae782"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TTMP2ELCNS2NE3U7BOG3JLLU3U/bundle.json","state_url":"https://pith.science/pith/TTMP2ELCNS2NE3U7BOG3JLLU3U/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TTMP2ELCNS2NE3U7BOG3JLLU3U/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":"2026-08-06T19:56:21Z","links":{"resolver":"https://pith.science/pith/TTMP2ELCNS2NE3U7BOG3JLLU3U","bundle":"https://pith.science/pith/TTMP2ELCNS2NE3U7BOG3JLLU3U/bundle.json","state":"https://pith.science/pith/TTMP2ELCNS2NE3U7BOG3JLLU3U/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TTMP2ELCNS2NE3U7BOG3JLLU3U/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:TTMP2ELCNS2NE3U7BOG3JLLU3U","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":"f4d84ce92e1adb54d3a9096a8d8106827bffb04d4949d4bdf4c395d1fe418287","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-02T08:42:33Z","title_canon_sha256":"08991a327db560e5a6db6adaf624791eb74809855484576470684b7a622370c2"},"schema_version":"1.0","source":{"id":"2106.01023","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.01023","created_at":"2026-07-05T02:45:42Z"},{"alias_kind":"arxiv_version","alias_value":"2106.01023v1","created_at":"2026-07-05T02:45:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.01023","created_at":"2026-07-05T02:45:42Z"},{"alias_kind":"pith_short_12","alias_value":"TTMP2ELCNS2N","created_at":"2026-07-05T02:45:42Z"},{"alias_kind":"pith_short_16","alias_value":"TTMP2ELCNS2NE3U7","created_at":"2026-07-05T02:45:42Z"},{"alias_kind":"pith_short_8","alias_value":"TTMP2ELC","created_at":"2026-07-05T02:45:42Z"}],"graph_snapshots":[{"event_id":"sha256:dcc673aedba14ded79a9525b23673da8c0f43c23285e12b32ebb0cf9bf8ae782","target":"graph","created_at":"2026-07-05T02:45:42Z","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/2106.01023/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pre-trained language models (PLMs) achieve great success in NLP. However, their huge model sizes hinder their applications in many practical systems. Knowledge distillation is a popular technique to compress PLMs, which learns a small student model from a large teacher PLM. However, the knowledge learned from a single teacher may be limited and even biased, resulting in low-quality student model. In this paper, we propose a multi-teacher knowledge distillation framework named MT-BERT for pre-trained language model compression, which can train high-quality student model from multiple teacher PL","authors_text":"Chuhan Wu, Fangzhao Wu, Yongfeng Huang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-02T08:42:33Z","title":"One Teacher is Enough? Pre-trained Language Model Distillation from Multiple Teachers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.01023","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:ef8a15e3d71722d61210c1b2d828f4e80b38ace5346b693ab1399f50fd75395d","target":"record","created_at":"2026-07-05T02:45:42Z","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":"f4d84ce92e1adb54d3a9096a8d8106827bffb04d4949d4bdf4c395d1fe418287","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-02T08:42:33Z","title_canon_sha256":"08991a327db560e5a6db6adaf624791eb74809855484576470684b7a622370c2"},"schema_version":"1.0","source":{"id":"2106.01023","kind":"arxiv","version":1}},"canonical_sha256":"9cd8fd11626cb4d26e9f0b8db4ad74dd27a9d08c19da3cf020f6048edaa3ab9f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9cd8fd11626cb4d26e9f0b8db4ad74dd27a9d08c19da3cf020f6048edaa3ab9f","first_computed_at":"2026-07-05T02:45:42.979614Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:45:42.979614Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"iV7/b90JucmE+fnnx9p5Rjq4aW8fOcx1Xp03goNslMh3eQfnMp9Rhw4/QC/CVa8VqtyJimoWL+3djnEtmesFDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T02:45:42.980039Z","signed_message":"canonical_sha256_bytes"},"source_id":"2106.01023","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ef8a15e3d71722d61210c1b2d828f4e80b38ace5346b693ab1399f50fd75395d","sha256:dcc673aedba14ded79a9525b23673da8c0f43c23285e12b32ebb0cf9bf8ae782"],"state_sha256":"f9a3337b86255805a3754124b38d3111d299cc8c0e57e9ccec8526f9b3bc1a47"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EXZBpzvJY0yxFWe5H67TFACW1XD9BDyLUt6ZunOHNkr86v/n/LO9SgvLcxpvxUjEmCHqTmtODIMlf7JheV0ABw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T19:56:21.840815Z","bundle_sha256":"17f25728d2b0094eb19fc20e1113f4ad7e6e77b3b6bbe4bb6ab26389ac84fd35"}}