{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:FBRGX74KI7XTPC2EOP5MYSK7QM","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":"a942c3fe4bc6663e8cbb867771f0d42b0c38bf9047b8303f827f8149d08101ea","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-15T14:10:58Z","title_canon_sha256":"887e0c248498c4c9de2aed8f5b3b5c092ebaa22477170bc92f6d85faf3ecadcd"},"schema_version":"1.0","source":{"id":"2604.13899","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.13899","created_at":"2026-06-19T16:11:23Z"},{"alias_kind":"arxiv_version","alias_value":"2604.13899v4","created_at":"2026-06-19T16:11:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.13899","created_at":"2026-06-19T16:11:23Z"},{"alias_kind":"pith_short_12","alias_value":"FBRGX74KI7XT","created_at":"2026-06-19T16:11:23Z"},{"alias_kind":"pith_short_16","alias_value":"FBRGX74KI7XTPC2E","created_at":"2026-06-19T16:11:23Z"},{"alias_kind":"pith_short_8","alias_value":"FBRGX74K","created_at":"2026-06-19T16:11:23Z"}],"graph_snapshots":[{"event_id":"sha256:1d6c22b580a4d6b364b3f018fe1c528fcd0e30658d1ffa3127f4b5442c81ad95","target":"graph","created_at":"2026-06-19T16:11:23Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"A classifier trained on 25,974 GPT-5.2 labels ($43) achieves comparable F1-Macro to one trained on 3,800 human annotations ($316)."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The assumption that the human-annotated subset serves as an unbiased gold standard and that the pre-enriched pool does not limit the potential benefits of active learning."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"LLM annotation can replace human labels for hostility detection with comparable F1 at much lower cost, but active learning adds little value and error structures differ systematically."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"LLM-generated labels train hostility detectors to the same F1-Macro level as human labels but at far lower cost."}],"snapshot_sha256":"c62b32ec2b4bd80f710c0eec62023310e2f5364db9b39ab0ba75633c4d3e603f"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.13899/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Instruction-tuned LLMs can annotate thousands of instances at low cost. This raises two questions for active learning (AL): can LLM labels replace human labels within the AL loop, and does AL remain necessary when entire corpora can be cheaply labeled? We investigate both on a new dataset of 277,902 German political TikTok comments (25,974 LLM-labeled, 5,000 human-annotated), comparing LLM and human annotation across seven conditions, four encoders, and 10 random seeds. Under a two-question interface that mirrors the human annotation task, LLM annotation at scale outperforms human-supervised c","authors_text":"Ahmad Dawar Hakimi, Hinrich Sch\\\"utze, Isabelle Augenstein, Lea Hirlimann","cross_cats":["cs.AI"],"headline":"LLM-generated labels train hostility detectors to the same F1-Macro level as human labels but at far lower cost.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-15T14:10:58Z","title":"Do We Still Need Humans in the Loop? Comparing Human and LLM Annotation in Active Learning for Hostility Detection"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.13899","kind":"arxiv","version":4},"verdict":{"created_at":"2026-05-10T13:33:05.188713Z","id":"b25afec2-5d9e-4081-9562-58568675e235","model_set":{"reader":"grok-4.3"},"one_line_summary":"LLM annotation can replace human labels for hostility detection with comparable F1 at much lower cost, but active learning adds little value and error structures differ systematically.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"LLM-generated labels train hostility detectors to the same F1-Macro level as human labels but at far lower cost.","strongest_claim":"A classifier trained on 25,974 GPT-5.2 labels ($43) achieves comparable F1-Macro to one trained on 3,800 human annotations ($316).","weakest_assumption":"The assumption that the human-annotated subset serves as an unbiased gold standard and that the pre-enriched pool does not limit the potential benefits of active learning."}},"verdict_id":"b25afec2-5d9e-4081-9562-58568675e235"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:750ea45c73044743b5316c5e16a8f949502e151c3c6aeca71d2ab83ae30949f5","target":"record","created_at":"2026-06-19T16:11:23Z","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":"a942c3fe4bc6663e8cbb867771f0d42b0c38bf9047b8303f827f8149d08101ea","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-15T14:10:58Z","title_canon_sha256":"887e0c248498c4c9de2aed8f5b3b5c092ebaa22477170bc92f6d85faf3ecadcd"},"schema_version":"1.0","source":{"id":"2604.13899","kind":"arxiv","version":4}},"canonical_sha256":"28626bff8a47ef378b4473facc495f832c2164cbb1bf19dced36bdebf3408918","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"28626bff8a47ef378b4473facc495f832c2164cbb1bf19dced36bdebf3408918","first_computed_at":"2026-06-19T16:11:23.065263Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:11:23.065263Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ARx6K/wK5TO0MISmoeNcWE93/IOZ6tm4hIpgCS2tUGoVEsLLoJN9Vgv628Mb/wk+Nt4qTpaWXs9vt6h2BnHSBg==","signature_status":"signed_v1","signed_at":"2026-06-19T16:11:23.065747Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.13899","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:750ea45c73044743b5316c5e16a8f949502e151c3c6aeca71d2ab83ae30949f5","sha256:1d6c22b580a4d6b364b3f018fe1c528fcd0e30658d1ffa3127f4b5442c81ad95"],"state_sha256":"09034cadd4a678692f39662a22fbecabd25e83558649ef39f9f7c301c7647ffa"}