{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:UYKZG27O3G2VRGBKDJB2CVE3YK","short_pith_number":"pith:UYKZG27O","canonical_record":{"source":{"id":"1906.08286","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-06-19T18:08:06Z","cross_cats_sorted":[],"title_canon_sha256":"e6a673760f08eec4d10d6a30fcd3ea94c4bbf92a8a1ce8008cd1aab74e97a660","abstract_canon_sha256":"b9b4020564c984e329bd0e87a2f7d31a48627f4de7631eac957d423c820f0114"},"schema_version":"1.0"},"canonical_sha256":"a615936beed9b558982a1a43a1549bc2bc0e449e6fe7ddb0593d1d43e307be2b","source":{"kind":"arxiv","id":"1906.08286","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1906.08286","created_at":"2026-05-17T23:42:52Z"},{"alias_kind":"arxiv_version","alias_value":"1906.08286v1","created_at":"2026-05-17T23:42:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.08286","created_at":"2026-05-17T23:42:52Z"},{"alias_kind":"pith_short_12","alias_value":"UYKZG27O3G2V","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_16","alias_value":"UYKZG27O3G2VRGBK","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_8","alias_value":"UYKZG27O","created_at":"2026-05-18T12:33:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:UYKZG27O3G2VRGBKDJB2CVE3YK","target":"record","payload":{"canonical_record":{"source":{"id":"1906.08286","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-06-19T18:08:06Z","cross_cats_sorted":[],"title_canon_sha256":"e6a673760f08eec4d10d6a30fcd3ea94c4bbf92a8a1ce8008cd1aab74e97a660","abstract_canon_sha256":"b9b4020564c984e329bd0e87a2f7d31a48627f4de7631eac957d423c820f0114"},"schema_version":"1.0"},"canonical_sha256":"a615936beed9b558982a1a43a1549bc2bc0e449e6fe7ddb0593d1d43e307be2b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:42:52.895626Z","signature_b64":"LdSN5kxRQ9shR+MNXG/WOC+sU44lIiBg2FPwwJubykooBtBPbOnlTQMloVWojXBEMj0pniDU6ZgsNAs+ReVnBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a615936beed9b558982a1a43a1549bc2bc0e449e6fe7ddb0593d1d43e307be2b","last_reissued_at":"2026-05-17T23:42:52.895143Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:42:52.895143Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1906.08286","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":"2026-05-17T23:42:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KLg9sJ1ZB4TBX/a8r7JABE5Tlly04ptCuDYz7fYUuSvICwC6UsnEyiif6E3cHItrEYTaHu+S1tdOWzG604joDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T22:49:48.915524Z"},"content_sha256":"b01fc74fea3f782d388c0f8e78f5b28a6cfac5257d9b3122e53e7c11aa78f3c3","schema_version":"1.0","event_id":"sha256:b01fc74fea3f782d388c0f8e78f5b28a6cfac5257d9b3122e53e7c11aa78f3c3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:UYKZG27O3G2VRGBKDJB2CVE3YK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Incorporating Priors with Feature Attribution on Text Classification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Besim Avci, Frederick Liu","submitted_at":"2019-06-19T18:08:06Z","abstract_excerpt":"Feature attribution methods, proposed recently, help users interpret the predictions of complex models. Our approach integrates feature attributions into the objective function to allow machine learning practitioners to incorporate priors in model building. To demonstrate the effectiveness our technique, we apply it to two tasks: (1) mitigating unintended bias in text classifiers by neutralizing identity terms; (2) improving classifier performance in a scarce data setting by forcing the model to focus on toxic terms. Our approach adds an L2 distance loss between feature attributions and task-s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.08286","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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:42:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L8QiH8bG+LGnQTzFLXZGbbzL4CDZVCCXb5kljIQs56UBAoqhBcaaVKVlfNjMpdkrqis6ZiMKoE2y7+ILWqQvAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T22:49:48.915882Z"},"content_sha256":"5a774cf8fd871decfaa807dd08839b838e567cef61a87458b1c41dea335a4038","schema_version":"1.0","event_id":"sha256:5a774cf8fd871decfaa807dd08839b838e567cef61a87458b1c41dea335a4038"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UYKZG27O3G2VRGBKDJB2CVE3YK/bundle.json","state_url":"https://pith.science/pith/UYKZG27O3G2VRGBKDJB2CVE3YK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UYKZG27O3G2VRGBKDJB2CVE3YK/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":"2026-05-27T22:49:48Z","links":{"resolver":"https://pith.science/pith/UYKZG27O3G2VRGBKDJB2CVE3YK","bundle":"https://pith.science/pith/UYKZG27O3G2VRGBKDJB2CVE3YK/bundle.json","state":"https://pith.science/pith/UYKZG27O3G2VRGBKDJB2CVE3YK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UYKZG27O3G2VRGBKDJB2CVE3YK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:UYKZG27O3G2VRGBKDJB2CVE3YK","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":"b9b4020564c984e329bd0e87a2f7d31a48627f4de7631eac957d423c820f0114","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-06-19T18:08:06Z","title_canon_sha256":"e6a673760f08eec4d10d6a30fcd3ea94c4bbf92a8a1ce8008cd1aab74e97a660"},"schema_version":"1.0","source":{"id":"1906.08286","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1906.08286","created_at":"2026-05-17T23:42:52Z"},{"alias_kind":"arxiv_version","alias_value":"1906.08286v1","created_at":"2026-05-17T23:42:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.08286","created_at":"2026-05-17T23:42:52Z"},{"alias_kind":"pith_short_12","alias_value":"UYKZG27O3G2V","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_16","alias_value":"UYKZG27O3G2VRGBK","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_8","alias_value":"UYKZG27O","created_at":"2026-05-18T12:33:30Z"}],"graph_snapshots":[{"event_id":"sha256:5a774cf8fd871decfaa807dd08839b838e567cef61a87458b1c41dea335a4038","target":"graph","created_at":"2026-05-17T23:42:52Z","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"},"paper":{"abstract_excerpt":"Feature attribution methods, proposed recently, help users interpret the predictions of complex models. Our approach integrates feature attributions into the objective function to allow machine learning practitioners to incorporate priors in model building. To demonstrate the effectiveness our technique, we apply it to two tasks: (1) mitigating unintended bias in text classifiers by neutralizing identity terms; (2) improving classifier performance in a scarce data setting by forcing the model to focus on toxic terms. Our approach adds an L2 distance loss between feature attributions and task-s","authors_text":"Besim Avci, Frederick Liu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-06-19T18:08:06Z","title":"Incorporating Priors with Feature Attribution on Text Classification"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.08286","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:b01fc74fea3f782d388c0f8e78f5b28a6cfac5257d9b3122e53e7c11aa78f3c3","target":"record","created_at":"2026-05-17T23:42:52Z","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":"b9b4020564c984e329bd0e87a2f7d31a48627f4de7631eac957d423c820f0114","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-06-19T18:08:06Z","title_canon_sha256":"e6a673760f08eec4d10d6a30fcd3ea94c4bbf92a8a1ce8008cd1aab74e97a660"},"schema_version":"1.0","source":{"id":"1906.08286","kind":"arxiv","version":1}},"canonical_sha256":"a615936beed9b558982a1a43a1549bc2bc0e449e6fe7ddb0593d1d43e307be2b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a615936beed9b558982a1a43a1549bc2bc0e449e6fe7ddb0593d1d43e307be2b","first_computed_at":"2026-05-17T23:42:52.895143Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:42:52.895143Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LdSN5kxRQ9shR+MNXG/WOC+sU44lIiBg2FPwwJubykooBtBPbOnlTQMloVWojXBEMj0pniDU6ZgsNAs+ReVnBA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:42:52.895626Z","signed_message":"canonical_sha256_bytes"},"source_id":"1906.08286","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b01fc74fea3f782d388c0f8e78f5b28a6cfac5257d9b3122e53e7c11aa78f3c3","sha256:5a774cf8fd871decfaa807dd08839b838e567cef61a87458b1c41dea335a4038"],"state_sha256":"71bba4cba2294771c3678f36257a0b8b10909587498059beb9cbd2e9a52e9eeb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CpTvFYjTS4n+VRu04uDrWJ0sqbM3lbSDNAReArrAhID8rdLrYhkC7u/2IymbxW9IjrId81By9gR08TD29CXgCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T22:49:48.918695Z","bundle_sha256":"c82f2826435edf5848e47468ad84f4d41577f6915efc008a6721a6c6836f09dc"}}