{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SOSNE3N2GEBFF47NNPDTNEVF26","short_pith_number":"pith:SOSNE3N2","schema_version":"1.0","canonical_sha256":"93a4d26dba310252f3ed6bc73692a5d784f929311b84cd0819fddf83ba4751e2","source":{"kind":"arxiv","id":"2412.08139","version":1},"attestation_state":"computed","paper":{"title":"Wasserstein Distance Rivals Kullback-Leibler Divergence for Knowledge Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Haoyuan Yang, Jiaming Lv, Peihua Li","submitted_at":"2024-12-11T06:54:39Z","abstract_excerpt":"Since pioneering work of Hinton et al., knowledge distillation based on Kullback-Leibler Divergence (KL-Div) has been predominant, and recently its variants have achieved compelling performance. However, KL-Div only compares probabilities of the corresponding category between the teacher and student while lacking a mechanism for cross-category comparison. Besides, KL-Div is problematic when applied to intermediate layers, as it cannot handle non-overlapping distributions and is unaware of geometry of the underlying manifold. To address these downsides, we propose a methodology of Wasserstein D"},"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":"2412.08139","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-11T06:54:39Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"e8a3a122a7271b11991364b812d08c6075ec9f8923536a29e62f3fd288e663c8","abstract_canon_sha256":"5e25aba6b043de0b9d1bc6bf3ab32c429053f7c334a494df241fe31e55d578bc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:47:44.926343Z","signature_b64":"cIfqtZHLwCHkMbtoAOz44086Z8xVX9FI+e+sDrsyy8AXvADVAX123aJjiqEX7MwPy/mr8ZdXNx+JMHApQ8ylCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"93a4d26dba310252f3ed6bc73692a5d784f929311b84cd0819fddf83ba4751e2","last_reissued_at":"2026-07-05T09:47:44.925925Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:47:44.925925Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Wasserstein Distance Rivals Kullback-Leibler Divergence for Knowledge Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Haoyuan Yang, Jiaming Lv, Peihua Li","submitted_at":"2024-12-11T06:54:39Z","abstract_excerpt":"Since pioneering work of Hinton et al., knowledge distillation based on Kullback-Leibler Divergence (KL-Div) has been predominant, and recently its variants have achieved compelling performance. However, KL-Div only compares probabilities of the corresponding category between the teacher and student while lacking a mechanism for cross-category comparison. Besides, KL-Div is problematic when applied to intermediate layers, as it cannot handle non-overlapping distributions and is unaware of geometry of the underlying manifold. To address these downsides, we propose a methodology of Wasserstein D"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.08139","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/2412.08139/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":"2412.08139","created_at":"2026-07-05T09:47:44.925986+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.08139v1","created_at":"2026-07-05T09:47:44.925986+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.08139","created_at":"2026-07-05T09:47:44.925986+00:00"},{"alias_kind":"pith_short_12","alias_value":"SOSNE3N2GEBF","created_at":"2026-07-05T09:47:44.925986+00:00"},{"alias_kind":"pith_short_16","alias_value":"SOSNE3N2GEBFF47N","created_at":"2026-07-05T09:47:44.925986+00:00"},{"alias_kind":"pith_short_8","alias_value":"SOSNE3N2","created_at":"2026-07-05T09:47:44.925986+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.01265","citing_title":"BicKD: Bilateral Contrastive Knowledge Distillation","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SOSNE3N2GEBFF47NNPDTNEVF26","json":"https://pith.science/pith/SOSNE3N2GEBFF47NNPDTNEVF26.json","graph_json":"https://pith.science/api/pith-number/SOSNE3N2GEBFF47NNPDTNEVF26/graph.json","events_json":"https://pith.science/api/pith-number/SOSNE3N2GEBFF47NNPDTNEVF26/events.json","paper":"https://pith.science/paper/SOSNE3N2"},"agent_actions":{"view_html":"https://pith.science/pith/SOSNE3N2GEBFF47NNPDTNEVF26","download_json":"https://pith.science/pith/SOSNE3N2GEBFF47NNPDTNEVF26.json","view_paper":"https://pith.science/paper/SOSNE3N2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.08139&json=true","fetch_graph":"https://pith.science/api/pith-number/SOSNE3N2GEBFF47NNPDTNEVF26/graph.json","fetch_events":"https://pith.science/api/pith-number/SOSNE3N2GEBFF47NNPDTNEVF26/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SOSNE3N2GEBFF47NNPDTNEVF26/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SOSNE3N2GEBFF47NNPDTNEVF26/action/storage_attestation","attest_author":"https://pith.science/pith/SOSNE3N2GEBFF47NNPDTNEVF26/action/author_attestation","sign_citation":"https://pith.science/pith/SOSNE3N2GEBFF47NNPDTNEVF26/action/citation_signature","submit_replication":"https://pith.science/pith/SOSNE3N2GEBFF47NNPDTNEVF26/action/replication_record"}},"created_at":"2026-07-05T09:47:44.925986+00:00","updated_at":"2026-07-05T09:47:44.925986+00:00"}