{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:23XEK7XJAGNMKSXOMI5NPWI5I3","short_pith_number":"pith:23XEK7XJ","schema_version":"1.0","canonical_sha256":"d6ee457ee9019ac54aee623ad7d91d46df1bbb7652cc0364f8dd9f8e4a5d3f6c","source":{"kind":"arxiv","id":"1812.07996","version":1},"attestation_state":"computed","paper":{"title":"Mining Interpretable AOG Representations from Convolutional Networks via Active Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Quanshi Zhang, Ruiming Cao, Song-Chun Zhu, Ying Nian Wu","submitted_at":"2018-12-18T05:49:36Z","abstract_excerpt":"In this paper, we present a method to mine object-part patterns from conv-layers of a pre-trained convolutional neural network (CNN). The mined object-part patterns are organized by an And-Or graph (AOG). This interpretable AOG representation consists of a four-layer semantic hierarchy, i.e., semantic parts, part templates, latent patterns, and neural units. The AOG associates each object part with certain neural units in feature maps of conv-layers. The AOG is constructed in a weakly-supervised manner, i.e., very few annotations (e.g., 3-20) of object parts are used to guide the learning of A"},"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":"1812.07996","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-12-18T05:49:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"fcf54b2f0785ecc565560b1586b878312be7d283930dd04513a2f95b785f25c4","abstract_canon_sha256":"aed2d6e9dd39e48c3b9a90285a6d876be43f706ed9d9537a3a7642e6e7f96158"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:57:55.224880Z","signature_b64":"KCmlMjxG+IP9oEjW0B4ynos9O7L9YeSwvbDGK1BzVhsOZodT9A7s28bunkkpRS5L5tqlLEDEfdrEtSqZfAwDBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d6ee457ee9019ac54aee623ad7d91d46df1bbb7652cc0364f8dd9f8e4a5d3f6c","last_reissued_at":"2026-05-17T23:57:55.224120Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:57:55.224120Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mining Interpretable AOG Representations from Convolutional Networks via Active Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Quanshi Zhang, Ruiming Cao, Song-Chun Zhu, Ying Nian Wu","submitted_at":"2018-12-18T05:49:36Z","abstract_excerpt":"In this paper, we present a method to mine object-part patterns from conv-layers of a pre-trained convolutional neural network (CNN). The mined object-part patterns are organized by an And-Or graph (AOG). This interpretable AOG representation consists of a four-layer semantic hierarchy, i.e., semantic parts, part templates, latent patterns, and neural units. The AOG associates each object part with certain neural units in feature maps of conv-layers. The AOG is constructed in a weakly-supervised manner, i.e., very few annotations (e.g., 3-20) of object parts are used to guide the learning of A"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1812.07996","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":""},"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":"1812.07996","created_at":"2026-05-17T23:57:55.224240+00:00"},{"alias_kind":"arxiv_version","alias_value":"1812.07996v1","created_at":"2026-05-17T23:57:55.224240+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1812.07996","created_at":"2026-05-17T23:57:55.224240+00:00"},{"alias_kind":"pith_short_12","alias_value":"23XEK7XJAGNM","created_at":"2026-05-18T12:31:59.375834+00:00"},{"alias_kind":"pith_short_16","alias_value":"23XEK7XJAGNMKSXO","created_at":"2026-05-18T12:31:59.375834+00:00"},{"alias_kind":"pith_short_8","alias_value":"23XEK7XJ","created_at":"2026-05-18T12:31:59.375834+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/23XEK7XJAGNMKSXOMI5NPWI5I3","json":"https://pith.science/pith/23XEK7XJAGNMKSXOMI5NPWI5I3.json","graph_json":"https://pith.science/api/pith-number/23XEK7XJAGNMKSXOMI5NPWI5I3/graph.json","events_json":"https://pith.science/api/pith-number/23XEK7XJAGNMKSXOMI5NPWI5I3/events.json","paper":"https://pith.science/paper/23XEK7XJ"},"agent_actions":{"view_html":"https://pith.science/pith/23XEK7XJAGNMKSXOMI5NPWI5I3","download_json":"https://pith.science/pith/23XEK7XJAGNMKSXOMI5NPWI5I3.json","view_paper":"https://pith.science/paper/23XEK7XJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1812.07996&json=true","fetch_graph":"https://pith.science/api/pith-number/23XEK7XJAGNMKSXOMI5NPWI5I3/graph.json","fetch_events":"https://pith.science/api/pith-number/23XEK7XJAGNMKSXOMI5NPWI5I3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/23XEK7XJAGNMKSXOMI5NPWI5I3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/23XEK7XJAGNMKSXOMI5NPWI5I3/action/storage_attestation","attest_author":"https://pith.science/pith/23XEK7XJAGNMKSXOMI5NPWI5I3/action/author_attestation","sign_citation":"https://pith.science/pith/23XEK7XJAGNMKSXOMI5NPWI5I3/action/citation_signature","submit_replication":"https://pith.science/pith/23XEK7XJAGNMKSXOMI5NPWI5I3/action/replication_record"}},"created_at":"2026-05-17T23:57:55.224240+00:00","updated_at":"2026-05-17T23:57:55.224240+00:00"}