{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:52SKZCPUKRXFTPRSTMRE6OCRTA","short_pith_number":"pith:52SKZCPU","schema_version":"1.0","canonical_sha256":"eea4ac89f4546e59be329b224f38519804a2e9bbe515d5a9e06f32a8713076db","source":{"kind":"arxiv","id":"2406.11275","version":2},"attestation_state":"computed","paper":{"title":"Self-training Large Language Models through Knowledge Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Erik Cambria, Przemyslaw Kazienko, Ranjan Satapathy, Teddy Ferdinan, Wei Jie Yeo","submitted_at":"2024-06-17T07:25:09Z","abstract_excerpt":"Large language models (LLMs) often necessitate extensive labeled datasets and training compute to achieve impressive performance across downstream tasks. This paper explores a self-training paradigm, where the LLM autonomously curates its own labels and selectively trains on unknown data samples identified through a reference-free consistency method. Empirical evaluations demonstrate significant improvements in reducing hallucination in generation across multiple subjects. Furthermore, the selective training framework mitigates catastrophic forgetting in out-of-distribution benchmarks, address"},"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":"2406.11275","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-17T07:25:09Z","cross_cats_sorted":[],"title_canon_sha256":"f39007266b033b1ae000776f67a3a51029fe1f46f705588cc7b922e715da639f","abstract_canon_sha256":"6b33e1049971e2efacbe2c41f0dbe95c8d21e3a4d5abf2e7d53d18c4215bd3a7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:34:14.964387Z","signature_b64":"qMBTZ6Ai5JO5z7qwCcv77maO/3VkAa7bAkkjp1Ek+I+n8xzb3b3PxvVskoHVhG7+Hy177Mi1h1KNK6G2+uMEBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eea4ac89f4546e59be329b224f38519804a2e9bbe515d5a9e06f32a8713076db","last_reissued_at":"2026-07-05T09:34:14.963930Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:34:14.963930Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-training Large Language Models through Knowledge Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Erik Cambria, Przemyslaw Kazienko, Ranjan Satapathy, Teddy Ferdinan, Wei Jie Yeo","submitted_at":"2024-06-17T07:25:09Z","abstract_excerpt":"Large language models (LLMs) often necessitate extensive labeled datasets and training compute to achieve impressive performance across downstream tasks. This paper explores a self-training paradigm, where the LLM autonomously curates its own labels and selectively trains on unknown data samples identified through a reference-free consistency method. Empirical evaluations demonstrate significant improvements in reducing hallucination in generation across multiple subjects. Furthermore, the selective training framework mitigates catastrophic forgetting in out-of-distribution benchmarks, address"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.11275","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/2406.11275/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":"2406.11275","created_at":"2026-07-05T09:34:14.963984+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.11275v2","created_at":"2026-07-05T09:34:14.963984+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.11275","created_at":"2026-07-05T09:34:14.963984+00:00"},{"alias_kind":"pith_short_12","alias_value":"52SKZCPUKRXF","created_at":"2026-07-05T09:34:14.963984+00:00"},{"alias_kind":"pith_short_16","alias_value":"52SKZCPUKRXFTPRS","created_at":"2026-07-05T09:34:14.963984+00:00"},{"alias_kind":"pith_short_8","alias_value":"52SKZCPU","created_at":"2026-07-05T09:34:14.963984+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.10906","citing_title":"SusGen-GPT: A Data-Centric LLM for Financial NLP and Sustainability Report Generation","ref_index":48,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/52SKZCPUKRXFTPRSTMRE6OCRTA","json":"https://pith.science/pith/52SKZCPUKRXFTPRSTMRE6OCRTA.json","graph_json":"https://pith.science/api/pith-number/52SKZCPUKRXFTPRSTMRE6OCRTA/graph.json","events_json":"https://pith.science/api/pith-number/52SKZCPUKRXFTPRSTMRE6OCRTA/events.json","paper":"https://pith.science/paper/52SKZCPU"},"agent_actions":{"view_html":"https://pith.science/pith/52SKZCPUKRXFTPRSTMRE6OCRTA","download_json":"https://pith.science/pith/52SKZCPUKRXFTPRSTMRE6OCRTA.json","view_paper":"https://pith.science/paper/52SKZCPU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.11275&json=true","fetch_graph":"https://pith.science/api/pith-number/52SKZCPUKRXFTPRSTMRE6OCRTA/graph.json","fetch_events":"https://pith.science/api/pith-number/52SKZCPUKRXFTPRSTMRE6OCRTA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/52SKZCPUKRXFTPRSTMRE6OCRTA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/52SKZCPUKRXFTPRSTMRE6OCRTA/action/storage_attestation","attest_author":"https://pith.science/pith/52SKZCPUKRXFTPRSTMRE6OCRTA/action/author_attestation","sign_citation":"https://pith.science/pith/52SKZCPUKRXFTPRSTMRE6OCRTA/action/citation_signature","submit_replication":"https://pith.science/pith/52SKZCPUKRXFTPRSTMRE6OCRTA/action/replication_record"}},"created_at":"2026-07-05T09:34:14.963984+00:00","updated_at":"2026-07-05T09:34:14.963984+00:00"}