{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ONBIPBWZNMS536ZLUBQ5IG2352","short_pith_number":"pith:ONBIPBWZ","schema_version":"1.0","canonical_sha256":"73428786d96b25ddfb2ba061d41b5beeb48e61a0734161b47765005a8b59a720","source":{"kind":"arxiv","id":"2112.01956","version":2},"attestation_state":"computed","paper":{"title":"Provably Valid and Diverse Mutations of Real-World Media Data for DNN Testing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.SE"],"primary_cat":"cs.LG","authors_text":"Qi Pang, Shuai Wang, Yuanyuan Yuan","submitted_at":"2021-12-03T15:02:22Z","abstract_excerpt":"Deep neural networks (DNNs) often accept high-dimensional media data (e.g., photos, text, and audio) and understand their perceptual content (e.g., a cat). To test DNNs, diverse inputs are needed to trigger mis-predictions. Some preliminary works use byte-level mutations or domain-specific filters (e.g., foggy), whose enabled mutations may be limited and likely error-prone. SOTA works employ deep generative models to generate (infinite) inputs. Also, to keep the mutated inputs perceptually valid (e.g., a cat remains a \"cat\" after mutation), existing efforts rely on imprecise and less generaliz"},"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":"2112.01956","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-12-03T15:02:22Z","cross_cats_sorted":["cs.CR","cs.SE"],"title_canon_sha256":"2d80fcbf04ba9b20282abb4dfd952ac5088ab7571828a7d64d7419d121690974","abstract_canon_sha256":"b2631c9bcb63576985a4400c07b7489eab45391a1ab1c10474c4458133ec4281"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:04:09.303182Z","signature_b64":"jQCHkpEUgwBBsCjXXuSh9+9UBNbxH2DB+PkF0dxk37LvIwN0uOt6TU8WR82Y/IzgbiWGcyMHVve06dH2JVLYDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"73428786d96b25ddfb2ba061d41b5beeb48e61a0734161b47765005a8b59a720","last_reissued_at":"2026-07-05T07:04:09.302693Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:04:09.302693Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Provably Valid and Diverse Mutations of Real-World Media Data for DNN Testing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.SE"],"primary_cat":"cs.LG","authors_text":"Qi Pang, Shuai Wang, Yuanyuan Yuan","submitted_at":"2021-12-03T15:02:22Z","abstract_excerpt":"Deep neural networks (DNNs) often accept high-dimensional media data (e.g., photos, text, and audio) and understand their perceptual content (e.g., a cat). To test DNNs, diverse inputs are needed to trigger mis-predictions. Some preliminary works use byte-level mutations or domain-specific filters (e.g., foggy), whose enabled mutations may be limited and likely error-prone. SOTA works employ deep generative models to generate (infinite) inputs. Also, to keep the mutated inputs perceptually valid (e.g., a cat remains a \"cat\" after mutation), existing efforts rely on imprecise and less generaliz"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.01956","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/2112.01956/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":"2112.01956","created_at":"2026-07-05T07:04:09.302750+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.01956v2","created_at":"2026-07-05T07:04:09.302750+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.01956","created_at":"2026-07-05T07:04:09.302750+00:00"},{"alias_kind":"pith_short_12","alias_value":"ONBIPBWZNMS5","created_at":"2026-07-05T07:04:09.302750+00:00"},{"alias_kind":"pith_short_16","alias_value":"ONBIPBWZNMS536ZL","created_at":"2026-07-05T07:04:09.302750+00:00"},{"alias_kind":"pith_short_8","alias_value":"ONBIPBWZ","created_at":"2026-07-05T07:04:09.302750+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/ONBIPBWZNMS536ZLUBQ5IG2352","json":"https://pith.science/pith/ONBIPBWZNMS536ZLUBQ5IG2352.json","graph_json":"https://pith.science/api/pith-number/ONBIPBWZNMS536ZLUBQ5IG2352/graph.json","events_json":"https://pith.science/api/pith-number/ONBIPBWZNMS536ZLUBQ5IG2352/events.json","paper":"https://pith.science/paper/ONBIPBWZ"},"agent_actions":{"view_html":"https://pith.science/pith/ONBIPBWZNMS536ZLUBQ5IG2352","download_json":"https://pith.science/pith/ONBIPBWZNMS536ZLUBQ5IG2352.json","view_paper":"https://pith.science/paper/ONBIPBWZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.01956&json=true","fetch_graph":"https://pith.science/api/pith-number/ONBIPBWZNMS536ZLUBQ5IG2352/graph.json","fetch_events":"https://pith.science/api/pith-number/ONBIPBWZNMS536ZLUBQ5IG2352/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ONBIPBWZNMS536ZLUBQ5IG2352/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ONBIPBWZNMS536ZLUBQ5IG2352/action/storage_attestation","attest_author":"https://pith.science/pith/ONBIPBWZNMS536ZLUBQ5IG2352/action/author_attestation","sign_citation":"https://pith.science/pith/ONBIPBWZNMS536ZLUBQ5IG2352/action/citation_signature","submit_replication":"https://pith.science/pith/ONBIPBWZNMS536ZLUBQ5IG2352/action/replication_record"}},"created_at":"2026-07-05T07:04:09.302750+00:00","updated_at":"2026-07-05T07:04:09.302750+00:00"}