{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:G4WP3TQ5PNKJLQQWQMLC2NGUAS","short_pith_number":"pith:G4WP3TQ5","schema_version":"1.0","canonical_sha256":"372cfdce1d7b5495c21683162d34d40496415a4f42f5215281dd3b00a2191124","source":{"kind":"arxiv","id":"2305.12870","version":2},"attestation_state":"computed","paper":{"title":"Lion: Adversarial Distillation of Proprietary Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chunkit Chan, Mingyang Chen, Wei Wang, Yuxin Jiang","submitted_at":"2023-05-22T09:49:16Z","abstract_excerpt":"The practice of transferring knowledge from a sophisticated, proprietary large language model (LLM) to a compact, open-source LLM has garnered considerable attention. Previous works have focused on a unidirectional knowledge distillation way by aligning the responses of the student model with those of the teacher model to a set of instructions. Nevertheless, they overlooked the possibility of incorporating any reciprocal \"feedback\"--identifying challenging instructions where the student model's performance falls short--to boost the student model's proficiency iteratively. To this end, we propo"},"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":"2305.12870","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-22T09:49:16Z","cross_cats_sorted":[],"title_canon_sha256":"14d9a3ff4154effa85f4f57274e3d014f16064c646ffdd7946071cd5f3a68207","abstract_canon_sha256":"56a669680359cbf5f548141ea130cd341d78baf303dd82cc195fa80b8e82b931"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:00:44.591746Z","signature_b64":"EBf74sPovXbjaIPNvaf1Fp6IglQ/gZIWN+ek7Y0i3C+B/fKVXZXUiyC4GZYi3QWk8EdkQ5X0QRxUdT1doBv9Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"372cfdce1d7b5495c21683162d34d40496415a4f42f5215281dd3b00a2191124","last_reissued_at":"2026-07-05T07:00:44.591272Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:00:44.591272Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Lion: Adversarial Distillation of Proprietary Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chunkit Chan, Mingyang Chen, Wei Wang, Yuxin Jiang","submitted_at":"2023-05-22T09:49:16Z","abstract_excerpt":"The practice of transferring knowledge from a sophisticated, proprietary large language model (LLM) to a compact, open-source LLM has garnered considerable attention. Previous works have focused on a unidirectional knowledge distillation way by aligning the responses of the student model with those of the teacher model to a set of instructions. Nevertheless, they overlooked the possibility of incorporating any reciprocal \"feedback\"--identifying challenging instructions where the student model's performance falls short--to boost the student model's proficiency iteratively. To this end, we propo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.12870","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/2305.12870/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":"2305.12870","created_at":"2026-07-05T07:00:44.591337+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.12870v2","created_at":"2026-07-05T07:00:44.591337+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.12870","created_at":"2026-07-05T07:00:44.591337+00:00"},{"alias_kind":"pith_short_12","alias_value":"G4WP3TQ5PNKJ","created_at":"2026-07-05T07:00:44.591337+00:00"},{"alias_kind":"pith_short_16","alias_value":"G4WP3TQ5PNKJLQQW","created_at":"2026-07-05T07:00:44.591337+00:00"},{"alias_kind":"pith_short_8","alias_value":"G4WP3TQ5","created_at":"2026-07-05T07:00:44.591337+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.14084","citing_title":"TIP: Token Importance in On-Policy Distillation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14084","citing_title":"TIP: Token Importance in On-Policy Distillation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2508.03949","citing_title":"Model Compression vs. Adversarial Robustness: An Empirical Study on Language Models for Code","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2511.05476","citing_title":"A Metamorphic Testing Perspective on Knowledge Distillation for Language Models of Code: Does the Student Deeply Mimic the Teacher?","ref_index":81,"is_internal_anchor":false},{"citing_arxiv_id":"2402.13116","citing_title":"A Survey on Knowledge Distillation of Large Language Models","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14084","citing_title":"TIP: Token Importance in On-Policy Distillation","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G4WP3TQ5PNKJLQQWQMLC2NGUAS","json":"https://pith.science/pith/G4WP3TQ5PNKJLQQWQMLC2NGUAS.json","graph_json":"https://pith.science/api/pith-number/G4WP3TQ5PNKJLQQWQMLC2NGUAS/graph.json","events_json":"https://pith.science/api/pith-number/G4WP3TQ5PNKJLQQWQMLC2NGUAS/events.json","paper":"https://pith.science/paper/G4WP3TQ5"},"agent_actions":{"view_html":"https://pith.science/pith/G4WP3TQ5PNKJLQQWQMLC2NGUAS","download_json":"https://pith.science/pith/G4WP3TQ5PNKJLQQWQMLC2NGUAS.json","view_paper":"https://pith.science/paper/G4WP3TQ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.12870&json=true","fetch_graph":"https://pith.science/api/pith-number/G4WP3TQ5PNKJLQQWQMLC2NGUAS/graph.json","fetch_events":"https://pith.science/api/pith-number/G4WP3TQ5PNKJLQQWQMLC2NGUAS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G4WP3TQ5PNKJLQQWQMLC2NGUAS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G4WP3TQ5PNKJLQQWQMLC2NGUAS/action/storage_attestation","attest_author":"https://pith.science/pith/G4WP3TQ5PNKJLQQWQMLC2NGUAS/action/author_attestation","sign_citation":"https://pith.science/pith/G4WP3TQ5PNKJLQQWQMLC2NGUAS/action/citation_signature","submit_replication":"https://pith.science/pith/G4WP3TQ5PNKJLQQWQMLC2NGUAS/action/replication_record"}},"created_at":"2026-07-05T07:00:44.591337+00:00","updated_at":"2026-07-05T07:00:44.591337+00:00"}