{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:HHCDRKGXXLAENG7TQJW73PSEGF","short_pith_number":"pith:HHCDRKGX","canonical_record":{"source":{"id":"2012.15761","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2020-12-31T17:36:48Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a9f7a8046e1fb889f3e542895a2d04fad264b6ea3544a653c8eda658d5b8b462","abstract_canon_sha256":"abd416482d598d47e322bc63763ab6ff165027e56203be60006a7c226c147255"},"schema_version":"1.0"},"canonical_sha256":"39c438a8d7bac0469bf3826dfdbe44317b878c78b66af0ab006965b573fc85c5","source":{"kind":"arxiv","id":"2012.15761","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2012.15761","created_at":"2026-07-05T02:45:49Z"},{"alias_kind":"arxiv_version","alias_value":"2012.15761v2","created_at":"2026-07-05T02:45:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.15761","created_at":"2026-07-05T02:45:49Z"},{"alias_kind":"pith_short_12","alias_value":"HHCDRKGXXLAE","created_at":"2026-07-05T02:45:49Z"},{"alias_kind":"pith_short_16","alias_value":"HHCDRKGXXLAENG7T","created_at":"2026-07-05T02:45:49Z"},{"alias_kind":"pith_short_8","alias_value":"HHCDRKGX","created_at":"2026-07-05T02:45:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:HHCDRKGXXLAENG7TQJW73PSEGF","target":"record","payload":{"canonical_record":{"source":{"id":"2012.15761","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2020-12-31T17:36:48Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a9f7a8046e1fb889f3e542895a2d04fad264b6ea3544a653c8eda658d5b8b462","abstract_canon_sha256":"abd416482d598d47e322bc63763ab6ff165027e56203be60006a7c226c147255"},"schema_version":"1.0"},"canonical_sha256":"39c438a8d7bac0469bf3826dfdbe44317b878c78b66af0ab006965b573fc85c5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:45:49.270066Z","signature_b64":"qXqIFu2GCRuA8FFyMpdXzstIghwwdfegHwVu/DuRAapP8bJItQpVVfQckmc1MBJON0AgR/O+QE2oWxGv1t5mBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"39c438a8d7bac0469bf3826dfdbe44317b878c78b66af0ab006965b573fc85c5","last_reissued_at":"2026-07-05T02:45:49.269587Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:45:49.269587Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2012.15761","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:45:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LJXrlicBingPtETcVSC8mwn898cAtWO133EmPca44lvivmUVZBp9pFCwBGbj5M1wmS/g9/LgRqgSDdTPVPrBCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T16:41:40.053848Z"},"content_sha256":"aef2219131e78f1e92030d83fa0baa93e8fd519c15f516c10b4f766bff2a8ba6","schema_version":"1.0","event_id":"sha256:aef2219131e78f1e92030d83fa0baa93e8fd519c15f516c10b4f766bff2a8ba6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:HHCDRKGXXLAENG7TQJW73PSEGF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning from the Worst: Dynamically Generated Datasets to Improve Online Hate Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Bertie Vidgen, Douwe Kiela, Tristan Thrush, Zeerak Waseem","submitted_at":"2020-12-31T17:36:48Z","abstract_excerpt":"We present a human-and-model-in-the-loop process for dynamically generating datasets and training better performing and more robust hate detection models. We provide a new dataset of ~40,000 entries, generated and labelled by trained annotators over four rounds of dynamic data creation. It includes ~15,000 challenging perturbations and each hateful entry has fine-grained labels for the type and target of hate. Hateful entries make up 54% of the dataset, which is substantially higher than comparable datasets. We show that model performance is substantially improved using this approach. Models t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.15761","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/2012.15761/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":"2026-07-05T02:45:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iFX6tvJEhbJkUj7AyMUUY7yBofoHz5HTTSVRQQlP05uD6wSoiIZ9DuNDkJV9DSV5W7W9CW2VYRx8cO7H5+N3DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T16:41:40.054365Z"},"content_sha256":"fa958a3e00fd8f14b98ec8d1f4b730068af9a2fe1d508fa33e77ee486d7d8276","schema_version":"1.0","event_id":"sha256:fa958a3e00fd8f14b98ec8d1f4b730068af9a2fe1d508fa33e77ee486d7d8276"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HHCDRKGXXLAENG7TQJW73PSEGF/bundle.json","state_url":"https://pith.science/pith/HHCDRKGXXLAENG7TQJW73PSEGF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HHCDRKGXXLAENG7TQJW73PSEGF/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":"2026-08-19T16:41:40Z","links":{"resolver":"https://pith.science/pith/HHCDRKGXXLAENG7TQJW73PSEGF","bundle":"https://pith.science/pith/HHCDRKGXXLAENG7TQJW73PSEGF/bundle.json","state":"https://pith.science/pith/HHCDRKGXXLAENG7TQJW73PSEGF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HHCDRKGXXLAENG7TQJW73PSEGF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:HHCDRKGXXLAENG7TQJW73PSEGF","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":"abd416482d598d47e322bc63763ab6ff165027e56203be60006a7c226c147255","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2020-12-31T17:36:48Z","title_canon_sha256":"a9f7a8046e1fb889f3e542895a2d04fad264b6ea3544a653c8eda658d5b8b462"},"schema_version":"1.0","source":{"id":"2012.15761","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2012.15761","created_at":"2026-07-05T02:45:49Z"},{"alias_kind":"arxiv_version","alias_value":"2012.15761v2","created_at":"2026-07-05T02:45:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.15761","created_at":"2026-07-05T02:45:49Z"},{"alias_kind":"pith_short_12","alias_value":"HHCDRKGXXLAE","created_at":"2026-07-05T02:45:49Z"},{"alias_kind":"pith_short_16","alias_value":"HHCDRKGXXLAENG7T","created_at":"2026-07-05T02:45:49Z"},{"alias_kind":"pith_short_8","alias_value":"HHCDRKGX","created_at":"2026-07-05T02:45:49Z"}],"graph_snapshots":[{"event_id":"sha256:fa958a3e00fd8f14b98ec8d1f4b730068af9a2fe1d508fa33e77ee486d7d8276","target":"graph","created_at":"2026-07-05T02:45:49Z","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/2012.15761/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present a human-and-model-in-the-loop process for dynamically generating datasets and training better performing and more robust hate detection models. We provide a new dataset of ~40,000 entries, generated and labelled by trained annotators over four rounds of dynamic data creation. It includes ~15,000 challenging perturbations and each hateful entry has fine-grained labels for the type and target of hate. Hateful entries make up 54% of the dataset, which is substantially higher than comparable datasets. We show that model performance is substantially improved using this approach. Models t","authors_text":"Bertie Vidgen, Douwe Kiela, Tristan Thrush, Zeerak Waseem","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2020-12-31T17:36:48Z","title":"Learning from the Worst: Dynamically Generated Datasets to Improve Online Hate Detection"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.15761","kind":"arxiv","version":2},"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:aef2219131e78f1e92030d83fa0baa93e8fd519c15f516c10b4f766bff2a8ba6","target":"record","created_at":"2026-07-05T02:45:49Z","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":"abd416482d598d47e322bc63763ab6ff165027e56203be60006a7c226c147255","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2020-12-31T17:36:48Z","title_canon_sha256":"a9f7a8046e1fb889f3e542895a2d04fad264b6ea3544a653c8eda658d5b8b462"},"schema_version":"1.0","source":{"id":"2012.15761","kind":"arxiv","version":2}},"canonical_sha256":"39c438a8d7bac0469bf3826dfdbe44317b878c78b66af0ab006965b573fc85c5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"39c438a8d7bac0469bf3826dfdbe44317b878c78b66af0ab006965b573fc85c5","first_computed_at":"2026-07-05T02:45:49.269587Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:45:49.269587Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qXqIFu2GCRuA8FFyMpdXzstIghwwdfegHwVu/DuRAapP8bJItQpVVfQckmc1MBJON0AgR/O+QE2oWxGv1t5mBA==","signature_status":"signed_v1","signed_at":"2026-07-05T02:45:49.270066Z","signed_message":"canonical_sha256_bytes"},"source_id":"2012.15761","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:aef2219131e78f1e92030d83fa0baa93e8fd519c15f516c10b4f766bff2a8ba6","sha256:fa958a3e00fd8f14b98ec8d1f4b730068af9a2fe1d508fa33e77ee486d7d8276"],"state_sha256":"c323e651887e8db3caa28198508cc9de4626f191e170e4cfda59c5b41e13bc15"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F5hlMJWoeita64zLJGOvx5SWHU0Zq/5ut1uC59QGTXaEI1SlbxrvMXG9OGmDX5f8fn8zJhx+Tys/8vXZPboKDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T16:41:40.058546Z","bundle_sha256":"bd9395a1e37ef2c26d2eee3ec6d61c4dc4240f6de252b5673d8a49f75c5764b2"}}