{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UJRYQESLAER4URISDGK4WLPTZE","short_pith_number":"pith:UJRYQESL","schema_version":"1.0","canonical_sha256":"a26388124b0123ca45121995cb2df3c92406b3229094622aff2111c3a283bdb2","source":{"kind":"arxiv","id":"2501.12332","version":1},"attestation_state":"computed","paper":{"title":"Automatic Labelling with Open-source LLMs using Dynamic Label Schema Integration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Chunyang Xiao, Fran Silavong, Sae Young Moon, Thomas Walshe, Yawwani Gunawardana","submitted_at":"2025-01-21T18:06:54Z","abstract_excerpt":"Acquiring labelled training data remains a costly task in real world machine learning projects to meet quantity and quality requirements. Recently Large Language Models (LLMs), notably GPT-4, have shown great promises in labelling data with high accuracy. However, privacy and cost concerns prevent the ubiquitous use of GPT-4. In this work, we explore effectively leveraging open-source models for automatic labelling. We identify integrating label schema as a promising technology but found that naively using the label description for classification leads to poor performance on high cardinality t"},"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":"2501.12332","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-21T18:06:54Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"665f6615023587acc5a277285a37e5277791681a09b639cf1f89cdcef5ef8b11","abstract_canon_sha256":"32ef96de54ddc3ff7c1fe1b1404d49c890f2ab59677531a9080a7569dd55fd6d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:03:34.706827Z","signature_b64":"sdYxNClJ+ILzVd2CrE6wcHrAhbWLyHTg2COtZPdZSWWiDXml9r8cIXQitSmFV/PTwp/Y2GWpoC8mvWe1AiwiBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a26388124b0123ca45121995cb2df3c92406b3229094622aff2111c3a283bdb2","last_reissued_at":"2026-07-05T10:03:34.706366Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:03:34.706366Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Automatic Labelling with Open-source LLMs using Dynamic Label Schema Integration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Chunyang Xiao, Fran Silavong, Sae Young Moon, Thomas Walshe, Yawwani Gunawardana","submitted_at":"2025-01-21T18:06:54Z","abstract_excerpt":"Acquiring labelled training data remains a costly task in real world machine learning projects to meet quantity and quality requirements. Recently Large Language Models (LLMs), notably GPT-4, have shown great promises in labelling data with high accuracy. However, privacy and cost concerns prevent the ubiquitous use of GPT-4. In this work, we explore effectively leveraging open-source models for automatic labelling. We identify integrating label schema as a promising technology but found that naively using the label description for classification leads to poor performance on high cardinality t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.12332","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2501.12332/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":"2501.12332","created_at":"2026-07-05T10:03:34.706437+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.12332v1","created_at":"2026-07-05T10:03:34.706437+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.12332","created_at":"2026-07-05T10:03:34.706437+00:00"},{"alias_kind":"pith_short_12","alias_value":"UJRYQESLAER4","created_at":"2026-07-05T10:03:34.706437+00:00"},{"alias_kind":"pith_short_16","alias_value":"UJRYQESLAER4URIS","created_at":"2026-07-05T10:03:34.706437+00:00"},{"alias_kind":"pith_short_8","alias_value":"UJRYQESL","created_at":"2026-07-05T10:03:34.706437+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.06813","citing_title":"A Comparative Benchmark of Large Language Models for Labelling Wind Turbine Maintenance Logs","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08628","citing_title":"Retrieval Augmented Classification for Confidential Documents","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UJRYQESLAER4URISDGK4WLPTZE","json":"https://pith.science/pith/UJRYQESLAER4URISDGK4WLPTZE.json","graph_json":"https://pith.science/api/pith-number/UJRYQESLAER4URISDGK4WLPTZE/graph.json","events_json":"https://pith.science/api/pith-number/UJRYQESLAER4URISDGK4WLPTZE/events.json","paper":"https://pith.science/paper/UJRYQESL"},"agent_actions":{"view_html":"https://pith.science/pith/UJRYQESLAER4URISDGK4WLPTZE","download_json":"https://pith.science/pith/UJRYQESLAER4URISDGK4WLPTZE.json","view_paper":"https://pith.science/paper/UJRYQESL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.12332&json=true","fetch_graph":"https://pith.science/api/pith-number/UJRYQESLAER4URISDGK4WLPTZE/graph.json","fetch_events":"https://pith.science/api/pith-number/UJRYQESLAER4URISDGK4WLPTZE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UJRYQESLAER4URISDGK4WLPTZE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UJRYQESLAER4URISDGK4WLPTZE/action/storage_attestation","attest_author":"https://pith.science/pith/UJRYQESLAER4URISDGK4WLPTZE/action/author_attestation","sign_citation":"https://pith.science/pith/UJRYQESLAER4URISDGK4WLPTZE/action/citation_signature","submit_replication":"https://pith.science/pith/UJRYQESLAER4URISDGK4WLPTZE/action/replication_record"}},"created_at":"2026-07-05T10:03:34.706437+00:00","updated_at":"2026-07-05T10:03:34.706437+00:00"}