{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:QQNC6Y6EEZPIKTFYFSOUN5LT6E","short_pith_number":"pith:QQNC6Y6E","schema_version":"1.0","canonical_sha256":"841a2f63c4265e854cb82c9d46f573f116e9a63ead3ef08676b375fec15e42f1","source":{"kind":"arxiv","id":"2210.03923","version":2},"attestation_state":"computed","paper":{"title":"Sparse Teachers Can Be Dense with Knowledge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chen Zhang, Dawei Song, Yi Yang","submitted_at":"2022-10-08T05:25:34Z","abstract_excerpt":"Recent advances in distilling pretrained language models have discovered that, besides the expressiveness of knowledge, the student-friendliness should be taken into consideration to realize a truly knowledgable teacher. Based on a pilot study, we find that over-parameterized teachers can produce expressive yet student-unfriendly knowledge and are thus limited in overall knowledgableness. To remove the parameters that result in student-unfriendliness, we propose a sparse teacher trick under the guidance of an overall knowledgable score for each teacher parameter. The knowledgable score is esse"},"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":"2210.03923","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-10-08T05:25:34Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"99d9004d226828c49b2785b787dad3ea5cf3a124fac75cd10d9e7b3a825647ba","abstract_canon_sha256":"459c67add8224031112038a9b4c0610d6de4f25ee3c9726b2c38a8a87bb78adc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:07:04.597254Z","signature_b64":"wY/QUm1ELfnM44R0F4cwkGXv0tHiPVbUCb2Zig7/LJBxN5AMl9re8Ug6t0Xyaxl+nMZrdSVL9W/MzOya/JC+DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"841a2f63c4265e854cb82c9d46f573f116e9a63ead3ef08676b375fec15e42f1","last_reissued_at":"2026-07-05T05:07:04.596904Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:07:04.596904Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sparse Teachers Can Be Dense with Knowledge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chen Zhang, Dawei Song, Yi Yang","submitted_at":"2022-10-08T05:25:34Z","abstract_excerpt":"Recent advances in distilling pretrained language models have discovered that, besides the expressiveness of knowledge, the student-friendliness should be taken into consideration to realize a truly knowledgable teacher. Based on a pilot study, we find that over-parameterized teachers can produce expressive yet student-unfriendly knowledge and are thus limited in overall knowledgableness. To remove the parameters that result in student-unfriendliness, we propose a sparse teacher trick under the guidance of an overall knowledgable score for each teacher parameter. The knowledgable score is esse"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.03923","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/2210.03923/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":"2210.03923","created_at":"2026-07-05T05:07:04.596961+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.03923v2","created_at":"2026-07-05T05:07:04.596961+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.03923","created_at":"2026-07-05T05:07:04.596961+00:00"},{"alias_kind":"pith_short_12","alias_value":"QQNC6Y6EEZPI","created_at":"2026-07-05T05:07:04.596961+00:00"},{"alias_kind":"pith_short_16","alias_value":"QQNC6Y6EEZPIKTFY","created_at":"2026-07-05T05:07:04.596961+00:00"},{"alias_kind":"pith_short_8","alias_value":"QQNC6Y6E","created_at":"2026-07-05T05:07:04.596961+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.04636","citing_title":"Put Teacher in Student's Shoes: Cross-Distillation for Ultra-compact Model Compression Framework","ref_index":48,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QQNC6Y6EEZPIKTFYFSOUN5LT6E","json":"https://pith.science/pith/QQNC6Y6EEZPIKTFYFSOUN5LT6E.json","graph_json":"https://pith.science/api/pith-number/QQNC6Y6EEZPIKTFYFSOUN5LT6E/graph.json","events_json":"https://pith.science/api/pith-number/QQNC6Y6EEZPIKTFYFSOUN5LT6E/events.json","paper":"https://pith.science/paper/QQNC6Y6E"},"agent_actions":{"view_html":"https://pith.science/pith/QQNC6Y6EEZPIKTFYFSOUN5LT6E","download_json":"https://pith.science/pith/QQNC6Y6EEZPIKTFYFSOUN5LT6E.json","view_paper":"https://pith.science/paper/QQNC6Y6E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.03923&json=true","fetch_graph":"https://pith.science/api/pith-number/QQNC6Y6EEZPIKTFYFSOUN5LT6E/graph.json","fetch_events":"https://pith.science/api/pith-number/QQNC6Y6EEZPIKTFYFSOUN5LT6E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QQNC6Y6EEZPIKTFYFSOUN5LT6E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QQNC6Y6EEZPIKTFYFSOUN5LT6E/action/storage_attestation","attest_author":"https://pith.science/pith/QQNC6Y6EEZPIKTFYFSOUN5LT6E/action/author_attestation","sign_citation":"https://pith.science/pith/QQNC6Y6EEZPIKTFYFSOUN5LT6E/action/citation_signature","submit_replication":"https://pith.science/pith/QQNC6Y6EEZPIKTFYFSOUN5LT6E/action/replication_record"}},"created_at":"2026-07-05T05:07:04.596961+00:00","updated_at":"2026-07-05T05:07:04.596961+00:00"}