{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FBRGX74KI7XTPC2EOP5MYSK7QM","short_pith_number":"pith:FBRGX74K","schema_version":"1.0","canonical_sha256":"28626bff8a47ef378b4473facc495f832c2164cbb1bf19dced36bdebf3408918","source":{"kind":"arxiv","id":"2604.13899","version":4},"attestation_state":"computed","paper":{"title":"Do We Still Need Humans in the Loop? Comparing Human and LLM Annotation in Active Learning for Hostility Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"LLM-generated labels train hostility detectors to the same F1-Macro level as human labels but at far lower cost.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ahmad Dawar Hakimi, Hinrich Sch\\\"utze, Isabelle Augenstein, Lea Hirlimann","submitted_at":"2026-04-15T14:10:58Z","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"},"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":"2604.13899","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-04-15T14:10:58Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"887e0c248498c4c9de2aed8f5b3b5c092ebaa22477170bc92f6d85faf3ecadcd","abstract_canon_sha256":"a942c3fe4bc6663e8cbb867771f0d42b0c38bf9047b8303f827f8149d08101ea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:11:23.065747Z","signature_b64":"ARx6K/wK5TO0MISmoeNcWE93/IOZ6tm4hIpgCS2tUGoVEsLLoJN9Vgv628Mb/wk+Nt4qTpaWXs9vt6h2BnHSBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"28626bff8a47ef378b4473facc495f832c2164cbb1bf19dced36bdebf3408918","last_reissued_at":"2026-06-19T16:11:23.065263Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:11:23.065263Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do We Still Need Humans in the Loop? Comparing Human and LLM Annotation in Active Learning for Hostility Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"LLM-generated labels train hostility detectors to the same F1-Macro level as human labels but at far lower cost.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ahmad Dawar Hakimi, Hinrich Sch\\\"utze, Isabelle Augenstein, Lea Hirlimann","submitted_at":"2026-04-15T14:10:58Z","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"},"claims":{"count":4,"items":[{"kind":"strongest_claim","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).","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","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.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","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.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"LLM-generated labels train hostility detectors to the same F1-Macro level as human labels but at far lower cost.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"c62b32ec2b4bd80f710c0eec62023310e2f5364db9b39ab0ba75633c4d3e603f"},"source":{"id":"2604.13899","kind":"arxiv","version":4},"verdict":{"id":"b25afec2-5d9e-4081-9562-58568675e235","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T13:33:05.188713Z","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).","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","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.","pith_extraction_headline":"LLM-generated labels train hostility detectors to the same F1-Macro level as human labels but at far lower cost."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.13899/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":"2604.13899","created_at":"2026-06-19T16:11:23.065355+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.13899v4","created_at":"2026-06-19T16:11:23.065355+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.13899","created_at":"2026-06-19T16:11:23.065355+00:00"},{"alias_kind":"pith_short_12","alias_value":"FBRGX74KI7XT","created_at":"2026-06-19T16:11:23.065355+00:00"},{"alias_kind":"pith_short_16","alias_value":"FBRGX74KI7XTPC2E","created_at":"2026-06-19T16:11:23.065355+00:00"},{"alias_kind":"pith_short_8","alias_value":"FBRGX74K","created_at":"2026-06-19T16:11:23.065355+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FBRGX74KI7XTPC2EOP5MYSK7QM","json":"https://pith.science/pith/FBRGX74KI7XTPC2EOP5MYSK7QM.json","graph_json":"https://pith.science/api/pith-number/FBRGX74KI7XTPC2EOP5MYSK7QM/graph.json","events_json":"https://pith.science/api/pith-number/FBRGX74KI7XTPC2EOP5MYSK7QM/events.json","paper":"https://pith.science/paper/FBRGX74K"},"agent_actions":{"view_html":"https://pith.science/pith/FBRGX74KI7XTPC2EOP5MYSK7QM","download_json":"https://pith.science/pith/FBRGX74KI7XTPC2EOP5MYSK7QM.json","view_paper":"https://pith.science/paper/FBRGX74K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.13899&json=true","fetch_graph":"https://pith.science/api/pith-number/FBRGX74KI7XTPC2EOP5MYSK7QM/graph.json","fetch_events":"https://pith.science/api/pith-number/FBRGX74KI7XTPC2EOP5MYSK7QM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FBRGX74KI7XTPC2EOP5MYSK7QM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FBRGX74KI7XTPC2EOP5MYSK7QM/action/storage_attestation","attest_author":"https://pith.science/pith/FBRGX74KI7XTPC2EOP5MYSK7QM/action/author_attestation","sign_citation":"https://pith.science/pith/FBRGX74KI7XTPC2EOP5MYSK7QM/action/citation_signature","submit_replication":"https://pith.science/pith/FBRGX74KI7XTPC2EOP5MYSK7QM/action/replication_record"}},"created_at":"2026-06-19T16:11:23.065355+00:00","updated_at":"2026-06-19T16:11:23.065355+00:00"}