{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:KQCORWKNFINP6R2OORU2LXH5JA","short_pith_number":"pith:KQCORWKN","canonical_record":{"source":{"id":"1911.06876","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-15T21:10:50Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"c8db0ab376a7f341e1100316fab9ada48b1ad9ea888f249f332fb74e398cfd58","abstract_canon_sha256":"35f253fda6b56fa5246a4c52d1c065f298dcb72ddfa882f24e5b648e9430993b"},"schema_version":"1.0"},"canonical_sha256":"5404e8d94d2a1aff474e7469a5dcfd481fe464e8f991b16e1725d44bf70b8bec","source":{"kind":"arxiv","id":"1911.06876","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1911.06876","created_at":"2026-07-05T00:19:40Z"},{"alias_kind":"arxiv_version","alias_value":"1911.06876v1","created_at":"2026-07-05T00:19:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.06876","created_at":"2026-07-05T00:19:40Z"},{"alias_kind":"pith_short_12","alias_value":"KQCORWKNFINP","created_at":"2026-07-05T00:19:40Z"},{"alias_kind":"pith_short_16","alias_value":"KQCORWKNFINP6R2O","created_at":"2026-07-05T00:19:40Z"},{"alias_kind":"pith_short_8","alias_value":"KQCORWKN","created_at":"2026-07-05T00:19:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:KQCORWKNFINP6R2OORU2LXH5JA","target":"record","payload":{"canonical_record":{"source":{"id":"1911.06876","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-15T21:10:50Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"c8db0ab376a7f341e1100316fab9ada48b1ad9ea888f249f332fb74e398cfd58","abstract_canon_sha256":"35f253fda6b56fa5246a4c52d1c065f298dcb72ddfa882f24e5b648e9430993b"},"schema_version":"1.0"},"canonical_sha256":"5404e8d94d2a1aff474e7469a5dcfd481fe464e8f991b16e1725d44bf70b8bec","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:19:40.238211Z","signature_b64":"LY+tz4JqqcA6gw+InbSBmABv1rqnyAFhzrLsplN4Z8+hZNZVghVkBTZlqlMClxe+Mv9YN3/OCAvkTzDWd0upCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5404e8d94d2a1aff474e7469a5dcfd481fe464e8f991b16e1725d44bf70b8bec","last_reissued_at":"2026-07-05T00:19:40.237758Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:19:40.237758Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1911.06876","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T00:19:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5WyjagHVcs+BN27ZS/m2cwH4JxKuJuQoEX4qsOlrbPSipK7mgiQr1xIffWxvZbuquMJkRx2z0xWPsZh5nWSQAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T21:23:22.982224Z"},"content_sha256":"b5b3340c2999c003df7c07c2fd27f9b82a8e7068132ae9cb7e812c665c8bb9dc","schema_version":"1.0","event_id":"sha256:b5b3340c2999c003df7c07c2fd27f9b82a8e7068132ae9cb7e812c665c8bb9dc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:KQCORWKNFINP6R2OORU2LXH5JA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Explanatory Masks for Neural Network Interpretability","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Garrett Goh, Lawrence Phillips, Nathan Hodas","submitted_at":"2019-11-15T21:10:50Z","abstract_excerpt":"Neural network interpretability is a vital component for applications across a wide variety of domains. In such cases it is often useful to analyze a network which has already been trained for its specific purpose. In this work, we develop a method to produce explanation masks for pre-trained networks. The mask localizes the most important aspects of each input for prediction of the original network. Masks are created by a secondary network whose goal is to create as small an explanation as possible while still preserving the predictive accuracy of the original network. We demonstrate the appl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.06876","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/1911.06876/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T00:19:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d4uojzkAqfPeZpf9KkTp2MTf9OvkF8DABLvE6g3zu2Z/ywmpVcNqOXwMr//Z+vwktBSA7YOHaQyuoi8QC3TaDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T21:23:22.982766Z"},"content_sha256":"b89ba30ebdba781d2f9ac856ef1a8ccfdfa862fef42130e3e17492d2caf86f2f","schema_version":"1.0","event_id":"sha256:b89ba30ebdba781d2f9ac856ef1a8ccfdfa862fef42130e3e17492d2caf86f2f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KQCORWKNFINP6R2OORU2LXH5JA/bundle.json","state_url":"https://pith.science/pith/KQCORWKNFINP6R2OORU2LXH5JA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KQCORWKNFINP6R2OORU2LXH5JA/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-14T21:23:22Z","links":{"resolver":"https://pith.science/pith/KQCORWKNFINP6R2OORU2LXH5JA","bundle":"https://pith.science/pith/KQCORWKNFINP6R2OORU2LXH5JA/bundle.json","state":"https://pith.science/pith/KQCORWKNFINP6R2OORU2LXH5JA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KQCORWKNFINP6R2OORU2LXH5JA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:KQCORWKNFINP6R2OORU2LXH5JA","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"35f253fda6b56fa5246a4c52d1c065f298dcb72ddfa882f24e5b648e9430993b","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-15T21:10:50Z","title_canon_sha256":"c8db0ab376a7f341e1100316fab9ada48b1ad9ea888f249f332fb74e398cfd58"},"schema_version":"1.0","source":{"id":"1911.06876","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1911.06876","created_at":"2026-07-05T00:19:40Z"},{"alias_kind":"arxiv_version","alias_value":"1911.06876v1","created_at":"2026-07-05T00:19:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.06876","created_at":"2026-07-05T00:19:40Z"},{"alias_kind":"pith_short_12","alias_value":"KQCORWKNFINP","created_at":"2026-07-05T00:19:40Z"},{"alias_kind":"pith_short_16","alias_value":"KQCORWKNFINP6R2O","created_at":"2026-07-05T00:19:40Z"},{"alias_kind":"pith_short_8","alias_value":"KQCORWKN","created_at":"2026-07-05T00:19:40Z"}],"graph_snapshots":[{"event_id":"sha256:b89ba30ebdba781d2f9ac856ef1a8ccfdfa862fef42130e3e17492d2caf86f2f","target":"graph","created_at":"2026-07-05T00:19:40Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/1911.06876/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Neural network interpretability is a vital component for applications across a wide variety of domains. In such cases it is often useful to analyze a network which has already been trained for its specific purpose. In this work, we develop a method to produce explanation masks for pre-trained networks. The mask localizes the most important aspects of each input for prediction of the original network. Masks are created by a secondary network whose goal is to create as small an explanation as possible while still preserving the predictive accuracy of the original network. We demonstrate the appl","authors_text":"Garrett Goh, Lawrence Phillips, Nathan Hodas","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-15T21:10:50Z","title":"Explanatory Masks for Neural Network Interpretability"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.06876","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b5b3340c2999c003df7c07c2fd27f9b82a8e7068132ae9cb7e812c665c8bb9dc","target":"record","created_at":"2026-07-05T00:19:40Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"35f253fda6b56fa5246a4c52d1c065f298dcb72ddfa882f24e5b648e9430993b","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-11-15T21:10:50Z","title_canon_sha256":"c8db0ab376a7f341e1100316fab9ada48b1ad9ea888f249f332fb74e398cfd58"},"schema_version":"1.0","source":{"id":"1911.06876","kind":"arxiv","version":1}},"canonical_sha256":"5404e8d94d2a1aff474e7469a5dcfd481fe464e8f991b16e1725d44bf70b8bec","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5404e8d94d2a1aff474e7469a5dcfd481fe464e8f991b16e1725d44bf70b8bec","first_computed_at":"2026-07-05T00:19:40.237758Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:19:40.237758Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LY+tz4JqqcA6gw+InbSBmABv1rqnyAFhzrLsplN4Z8+hZNZVghVkBTZlqlMClxe+Mv9YN3/OCAvkTzDWd0upCw==","signature_status":"signed_v1","signed_at":"2026-07-05T00:19:40.238211Z","signed_message":"canonical_sha256_bytes"},"source_id":"1911.06876","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b5b3340c2999c003df7c07c2fd27f9b82a8e7068132ae9cb7e812c665c8bb9dc","sha256:b89ba30ebdba781d2f9ac856ef1a8ccfdfa862fef42130e3e17492d2caf86f2f"],"state_sha256":"e28a5d1c6b88a38bf0b1317e73fc6ffba9bcf1f1e1ddaa3e3720b379c0e1d163"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PzkZd+b1lZrR/aNaBmfexjNrnVAAalb9sNsW+Sfrm4fTsHBDYmY68eGeqzeMVCMq70zdCEMJ+5jKx8vUtpZKCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T21:23:22.990228Z","bundle_sha256":"6072de9afd8cdc158249689ee01bfd9051c4e58eea8ef4750b3e0ce6da82c8c1"}}