{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FTJMDNAKIAIK3CKEMK72YM3J7H","short_pith_number":"pith:FTJMDNAK","schema_version":"1.0","canonical_sha256":"2cd2c1b40a4010ad894462bfac3369f9f728ac1f54ee869ad58886c8e0972ca9","source":{"kind":"arxiv","id":"2501.17840","version":1},"attestation_state":"computed","paper":{"title":"Learning Beyond the Surface: How Far Can Continual Pre-Training with LoRA Enhance LLMs' Domain-Specific Insight Learning?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Estevam Hruschka, Pouya Pezeshkpour","submitted_at":"2025-01-29T18:40:32Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated remarkable performance on various tasks, yet their ability to extract and internalize deeper insights from domain-specific datasets remains underexplored. In this study, we investigate how continual pre-training can enhance LLMs' capacity for insight learning across three distinct forms: declarative, statistical, and probabilistic insights. Focusing on two critical domains: medicine and finance, we employ LoRA to train LLMs on two existing datasets. To evaluate each insight type, we create benchmarks to measure how well continual pre-training help"},"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":"2501.17840","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-29T18:40:32Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"62761c98d5c427acb88e90913567a4c01f9f236c00eed81f5251f861fe65d497","abstract_canon_sha256":"467a3341b09336b6d2db434d54431059e82f537d9362de07b8cf9f1716f8dcbc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:06:58.875343Z","signature_b64":"gTJPjkfW+FdD+4/CFipaks4ee2FeHVI7ctNPphnt9IhJ9n5d7kzbAP7bVH/HgWN5iN+EBqAQ4e4CEVwGsHXSAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2cd2c1b40a4010ad894462bfac3369f9f728ac1f54ee869ad58886c8e0972ca9","last_reissued_at":"2026-07-05T10:06:58.874910Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:06:58.874910Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Beyond the Surface: How Far Can Continual Pre-Training with LoRA Enhance LLMs' Domain-Specific Insight Learning?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Estevam Hruschka, Pouya Pezeshkpour","submitted_at":"2025-01-29T18:40:32Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated remarkable performance on various tasks, yet their ability to extract and internalize deeper insights from domain-specific datasets remains underexplored. In this study, we investigate how continual pre-training can enhance LLMs' capacity for insight learning across three distinct forms: declarative, statistical, and probabilistic insights. Focusing on two critical domains: medicine and finance, we employ LoRA to train LLMs on two existing datasets. To evaluate each insight type, we create benchmarks to measure how well continual pre-training help"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.17840","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/2501.17840/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":"2501.17840","created_at":"2026-07-05T10:06:58.874966+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.17840v1","created_at":"2026-07-05T10:06:58.874966+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.17840","created_at":"2026-07-05T10:06:58.874966+00:00"},{"alias_kind":"pith_short_12","alias_value":"FTJMDNAKIAIK","created_at":"2026-07-05T10:06:58.874966+00:00"},{"alias_kind":"pith_short_16","alias_value":"FTJMDNAKIAIK3CKE","created_at":"2026-07-05T10:06:58.874966+00:00"},{"alias_kind":"pith_short_8","alias_value":"FTJMDNAK","created_at":"2026-07-05T10:06:58.874966+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20359","citing_title":"Train, Retrieve, or Both? A Four-Arm Head-to-Head for Correct Statutory Citation on the Ontario Residential Tenancies Act","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FTJMDNAKIAIK3CKEMK72YM3J7H","json":"https://pith.science/pith/FTJMDNAKIAIK3CKEMK72YM3J7H.json","graph_json":"https://pith.science/api/pith-number/FTJMDNAKIAIK3CKEMK72YM3J7H/graph.json","events_json":"https://pith.science/api/pith-number/FTJMDNAKIAIK3CKEMK72YM3J7H/events.json","paper":"https://pith.science/paper/FTJMDNAK"},"agent_actions":{"view_html":"https://pith.science/pith/FTJMDNAKIAIK3CKEMK72YM3J7H","download_json":"https://pith.science/pith/FTJMDNAKIAIK3CKEMK72YM3J7H.json","view_paper":"https://pith.science/paper/FTJMDNAK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.17840&json=true","fetch_graph":"https://pith.science/api/pith-number/FTJMDNAKIAIK3CKEMK72YM3J7H/graph.json","fetch_events":"https://pith.science/api/pith-number/FTJMDNAKIAIK3CKEMK72YM3J7H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FTJMDNAKIAIK3CKEMK72YM3J7H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FTJMDNAKIAIK3CKEMK72YM3J7H/action/storage_attestation","attest_author":"https://pith.science/pith/FTJMDNAKIAIK3CKEMK72YM3J7H/action/author_attestation","sign_citation":"https://pith.science/pith/FTJMDNAKIAIK3CKEMK72YM3J7H/action/citation_signature","submit_replication":"https://pith.science/pith/FTJMDNAKIAIK3CKEMK72YM3J7H/action/replication_record"}},"created_at":"2026-07-05T10:06:58.874966+00:00","updated_at":"2026-07-05T10:06:58.874966+00:00"}