{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:ENRZIHXWTUE2BOPGXGLVXZX7F2","short_pith_number":"pith:ENRZIHXW","schema_version":"1.0","canonical_sha256":"2363941ef69d09a0b9e6b9975be6ff2ebd8829656e3bc36127c35e3850cfba0f","source":{"kind":"arxiv","id":"1910.02673","version":1},"attestation_state":"computed","paper":{"title":"Interpretable Disentanglement of Neural Networks by Extracting Class-Specific Subnetwork","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Hang Su, Xiaolin Hu, Yulong Wang","submitted_at":"2019-10-07T08:42:45Z","abstract_excerpt":"We propose a novel perspective to understand deep neural networks in an interpretable disentanglement form. For each semantic class, we extract a class-specific functional subnetwork from the original full model, with compressed structure while maintaining comparable prediction performance. The structure representations of extracted subnetworks display a resemblance to their corresponding class semantic similarities. We also apply extracted subnetworks in visual explanation and adversarial example detection tasks by merely replacing the original full model with class-specific subnetworks. Expe"},"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":"1910.02673","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2019-10-07T08:42:45Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"50f88632a637ceb0268c966185e44a5180eb996a088043d744ddbac8f8857680","abstract_canon_sha256":"823d2a54469253e5b74eed91b8279b26d1673e2c5372ff7bc31999a356f18c12"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:10:10.395304Z","signature_b64":"WZHa+DOSx4JlZyORybOidVwjrBuvxKjtfcLiLsJ4cy6gxE1RTB1MuTEoL6FclNr6d8yu8ELM6cQFrcV3xhi2Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2363941ef69d09a0b9e6b9975be6ff2ebd8829656e3bc36127c35e3850cfba0f","last_reissued_at":"2026-07-05T00:10:10.394958Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:10:10.394958Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Interpretable Disentanglement of Neural Networks by Extracting Class-Specific Subnetwork","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Hang Su, Xiaolin Hu, Yulong Wang","submitted_at":"2019-10-07T08:42:45Z","abstract_excerpt":"We propose a novel perspective to understand deep neural networks in an interpretable disentanglement form. For each semantic class, we extract a class-specific functional subnetwork from the original full model, with compressed structure while maintaining comparable prediction performance. The structure representations of extracted subnetworks display a resemblance to their corresponding class semantic similarities. We also apply extracted subnetworks in visual explanation and adversarial example detection tasks by merely replacing the original full model with class-specific subnetworks. Expe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.02673","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/1910.02673/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":"1910.02673","created_at":"2026-07-05T00:10:10.395015+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.02673v1","created_at":"2026-07-05T00:10:10.395015+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.02673","created_at":"2026-07-05T00:10:10.395015+00:00"},{"alias_kind":"pith_short_12","alias_value":"ENRZIHXWTUE2","created_at":"2026-07-05T00:10:10.395015+00:00"},{"alias_kind":"pith_short_16","alias_value":"ENRZIHXWTUE2BOPG","created_at":"2026-07-05T00:10:10.395015+00:00"},{"alias_kind":"pith_short_8","alias_value":"ENRZIHXW","created_at":"2026-07-05T00:10:10.395015+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ENRZIHXWTUE2BOPGXGLVXZX7F2","json":"https://pith.science/pith/ENRZIHXWTUE2BOPGXGLVXZX7F2.json","graph_json":"https://pith.science/api/pith-number/ENRZIHXWTUE2BOPGXGLVXZX7F2/graph.json","events_json":"https://pith.science/api/pith-number/ENRZIHXWTUE2BOPGXGLVXZX7F2/events.json","paper":"https://pith.science/paper/ENRZIHXW"},"agent_actions":{"view_html":"https://pith.science/pith/ENRZIHXWTUE2BOPGXGLVXZX7F2","download_json":"https://pith.science/pith/ENRZIHXWTUE2BOPGXGLVXZX7F2.json","view_paper":"https://pith.science/paper/ENRZIHXW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.02673&json=true","fetch_graph":"https://pith.science/api/pith-number/ENRZIHXWTUE2BOPGXGLVXZX7F2/graph.json","fetch_events":"https://pith.science/api/pith-number/ENRZIHXWTUE2BOPGXGLVXZX7F2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ENRZIHXWTUE2BOPGXGLVXZX7F2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ENRZIHXWTUE2BOPGXGLVXZX7F2/action/storage_attestation","attest_author":"https://pith.science/pith/ENRZIHXWTUE2BOPGXGLVXZX7F2/action/author_attestation","sign_citation":"https://pith.science/pith/ENRZIHXWTUE2BOPGXGLVXZX7F2/action/citation_signature","submit_replication":"https://pith.science/pith/ENRZIHXWTUE2BOPGXGLVXZX7F2/action/replication_record"}},"created_at":"2026-07-05T00:10:10.395015+00:00","updated_at":"2026-07-05T00:10:10.395015+00:00"}