{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:UFDBOHJBR4FY5ZC2CFQ2WFJFRN","short_pith_number":"pith:UFDBOHJB","schema_version":"1.0","canonical_sha256":"a146171d218f0b8ee45a1161ab15258b710d080697d9c485786ca33d276860f6","source":{"kind":"arxiv","id":"2304.13861","version":2},"attestation_state":"computed","paper":{"title":"The Parrot Dilemma: Human-Labeled vs. LLM-augmented Data in Classification Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CY","physics.soc-ph"],"primary_cat":"cs.CL","authors_text":"Anders Giovanni M{\\o}ller, Arianna Pera, Jacob Aarup Dalsgaard, Luca Maria Aiello","submitted_at":"2023-04-26T23:09:02Z","abstract_excerpt":"In the realm of Computational Social Science (CSS), practitioners often navigate complex, low-resource domains and face the costly and time-intensive challenges of acquiring and annotating data. We aim to establish a set of guidelines to address such challenges, comparing the use of human-labeled data with synthetically generated data from GPT-4 and Llama-2 in ten distinct CSS classification tasks of varying complexity. Additionally, we examine the impact of training data sizes on performance. Our findings reveal that models trained on human-labeled data consistently exhibit superior or compar"},"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":"2304.13861","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-04-26T23:09:02Z","cross_cats_sorted":["cs.CY","physics.soc-ph"],"title_canon_sha256":"99adc1204709beacd15e4a92244e0cc417c7fcae485ccef2a8836ab797d31331","abstract_canon_sha256":"b85509dd861ff17f62025204c5a9264e411a82a77c5db3786f0841915e9ca342"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:41:26.680841Z","signature_b64":"ibWQ9piugcf6JSrbALEHWDH4S3lpiJB71lODuqwTaxH35onOaxKAcki+6IDduc6IFLrnF96SCnblFTYy6+/rCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a146171d218f0b8ee45a1161ab15258b710d080697d9c485786ca33d276860f6","last_reissued_at":"2026-07-05T07:41:26.680443Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:41:26.680443Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Parrot Dilemma: Human-Labeled vs. LLM-augmented Data in Classification Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CY","physics.soc-ph"],"primary_cat":"cs.CL","authors_text":"Anders Giovanni M{\\o}ller, Arianna Pera, Jacob Aarup Dalsgaard, Luca Maria Aiello","submitted_at":"2023-04-26T23:09:02Z","abstract_excerpt":"In the realm of Computational Social Science (CSS), practitioners often navigate complex, low-resource domains and face the costly and time-intensive challenges of acquiring and annotating data. We aim to establish a set of guidelines to address such challenges, comparing the use of human-labeled data with synthetically generated data from GPT-4 and Llama-2 in ten distinct CSS classification tasks of varying complexity. Additionally, we examine the impact of training data sizes on performance. Our findings reveal that models trained on human-labeled data consistently exhibit superior or compar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.13861","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/2304.13861/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":"2304.13861","created_at":"2026-07-05T07:41:26.680500+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.13861v2","created_at":"2026-07-05T07:41:26.680500+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.13861","created_at":"2026-07-05T07:41:26.680500+00:00"},{"alias_kind":"pith_short_12","alias_value":"UFDBOHJBR4FY","created_at":"2026-07-05T07:41:26.680500+00:00"},{"alias_kind":"pith_short_16","alias_value":"UFDBOHJBR4FY5ZC2","created_at":"2026-07-05T07:41:26.680500+00:00"},{"alias_kind":"pith_short_8","alias_value":"UFDBOHJB","created_at":"2026-07-05T07:41:26.680500+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/UFDBOHJBR4FY5ZC2CFQ2WFJFRN","json":"https://pith.science/pith/UFDBOHJBR4FY5ZC2CFQ2WFJFRN.json","graph_json":"https://pith.science/api/pith-number/UFDBOHJBR4FY5ZC2CFQ2WFJFRN/graph.json","events_json":"https://pith.science/api/pith-number/UFDBOHJBR4FY5ZC2CFQ2WFJFRN/events.json","paper":"https://pith.science/paper/UFDBOHJB"},"agent_actions":{"view_html":"https://pith.science/pith/UFDBOHJBR4FY5ZC2CFQ2WFJFRN","download_json":"https://pith.science/pith/UFDBOHJBR4FY5ZC2CFQ2WFJFRN.json","view_paper":"https://pith.science/paper/UFDBOHJB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.13861&json=true","fetch_graph":"https://pith.science/api/pith-number/UFDBOHJBR4FY5ZC2CFQ2WFJFRN/graph.json","fetch_events":"https://pith.science/api/pith-number/UFDBOHJBR4FY5ZC2CFQ2WFJFRN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UFDBOHJBR4FY5ZC2CFQ2WFJFRN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UFDBOHJBR4FY5ZC2CFQ2WFJFRN/action/storage_attestation","attest_author":"https://pith.science/pith/UFDBOHJBR4FY5ZC2CFQ2WFJFRN/action/author_attestation","sign_citation":"https://pith.science/pith/UFDBOHJBR4FY5ZC2CFQ2WFJFRN/action/citation_signature","submit_replication":"https://pith.science/pith/UFDBOHJBR4FY5ZC2CFQ2WFJFRN/action/replication_record"}},"created_at":"2026-07-05T07:41:26.680500+00:00","updated_at":"2026-07-05T07:41:26.680500+00:00"}