{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:LBKJCLYXZDM4SODIIQQVFHI2VF","short_pith_number":"pith:LBKJCLYX","schema_version":"1.0","canonical_sha256":"5854912f17c8d9c938684421529d1aa95902f4e5fbb0caaec7a24937a3a4a435","source":{"kind":"arxiv","id":"2104.14435","version":2},"attestation_state":"computed","paper":{"title":"Customizable Reference Runtime Monitoring of Neural Networks using Resolution Boxes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Changshun Wu, Saddek Bensalem, Yli\\`es Falcone","submitted_at":"2021-04-25T21:58:02Z","abstract_excerpt":"Classification neural networks fail to detect inputs that do not fall inside the classes they have been trained for. Runtime monitoring techniques on the neuron activation pattern can be used to detect such inputs. We present an approach for monitoring classification systems via data abstraction. Data abstraction relies on the notion of box with a resolution. Box-based abstraction consists in representing a set of values by its minimal and maximal values in each dimension. We augment boxes with a notion of resolution and define their clustering coverage, which is intuitively a quantitative met"},"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":"2104.14435","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-04-25T21:58:02Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"386d1187996dbd95a4ca8f28a87f0530f92a8c366d5bb913dcd19f39223c6d9a","abstract_canon_sha256":"d5bbb2c2b4a6244ff9b86c602ba31d55171a21820adc4798af2331d45427622c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:56:58.058843Z","signature_b64":"hr1x3huw0dNk6ND3i378IUuM8Lc4liaTpJqnJaSdT8RebaQPxDHyc32lDmgGh565LqliN1lL25Ju8N5SwFJLDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5854912f17c8d9c938684421529d1aa95902f4e5fbb0caaec7a24937a3a4a435","last_reissued_at":"2026-07-05T02:56:58.058343Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:56:58.058343Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Customizable Reference Runtime Monitoring of Neural Networks using Resolution Boxes","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Changshun Wu, Saddek Bensalem, Yli\\`es Falcone","submitted_at":"2021-04-25T21:58:02Z","abstract_excerpt":"Classification neural networks fail to detect inputs that do not fall inside the classes they have been trained for. Runtime monitoring techniques on the neuron activation pattern can be used to detect such inputs. We present an approach for monitoring classification systems via data abstraction. Data abstraction relies on the notion of box with a resolution. Box-based abstraction consists in representing a set of values by its minimal and maximal values in each dimension. We augment boxes with a notion of resolution and define their clustering coverage, which is intuitively a quantitative met"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.14435","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/2104.14435/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":"2104.14435","created_at":"2026-07-05T02:56:58.058413+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.14435v2","created_at":"2026-07-05T02:56:58.058413+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.14435","created_at":"2026-07-05T02:56:58.058413+00:00"},{"alias_kind":"pith_short_12","alias_value":"LBKJCLYXZDM4","created_at":"2026-07-05T02:56:58.058413+00:00"},{"alias_kind":"pith_short_16","alias_value":"LBKJCLYXZDM4SODI","created_at":"2026-07-05T02:56:58.058413+00:00"},{"alias_kind":"pith_short_8","alias_value":"LBKJCLYX","created_at":"2026-07-05T02:56:58.058413+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.06869","citing_title":"Safety Monitoring of Machine Learning Perception Functions: a Survey","ref_index":96,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LBKJCLYXZDM4SODIIQQVFHI2VF","json":"https://pith.science/pith/LBKJCLYXZDM4SODIIQQVFHI2VF.json","graph_json":"https://pith.science/api/pith-number/LBKJCLYXZDM4SODIIQQVFHI2VF/graph.json","events_json":"https://pith.science/api/pith-number/LBKJCLYXZDM4SODIIQQVFHI2VF/events.json","paper":"https://pith.science/paper/LBKJCLYX"},"agent_actions":{"view_html":"https://pith.science/pith/LBKJCLYXZDM4SODIIQQVFHI2VF","download_json":"https://pith.science/pith/LBKJCLYXZDM4SODIIQQVFHI2VF.json","view_paper":"https://pith.science/paper/LBKJCLYX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.14435&json=true","fetch_graph":"https://pith.science/api/pith-number/LBKJCLYXZDM4SODIIQQVFHI2VF/graph.json","fetch_events":"https://pith.science/api/pith-number/LBKJCLYXZDM4SODIIQQVFHI2VF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LBKJCLYXZDM4SODIIQQVFHI2VF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LBKJCLYXZDM4SODIIQQVFHI2VF/action/storage_attestation","attest_author":"https://pith.science/pith/LBKJCLYXZDM4SODIIQQVFHI2VF/action/author_attestation","sign_citation":"https://pith.science/pith/LBKJCLYXZDM4SODIIQQVFHI2VF/action/citation_signature","submit_replication":"https://pith.science/pith/LBKJCLYXZDM4SODIIQQVFHI2VF/action/replication_record"}},"created_at":"2026-07-05T02:56:58.058413+00:00","updated_at":"2026-07-05T02:56:58.058413+00:00"}