{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:POJXT5BPBFIHDLFGVHZ6QRGLSJ","short_pith_number":"pith:POJXT5BP","schema_version":"1.0","canonical_sha256":"7b9379f42f095071aca6a9f3e844cb9273771717d57bb0a5bd98440d3bbe9af1","source":{"kind":"arxiv","id":"2603.14463","version":2},"attestation_state":"computed","paper":{"title":"An Industrial-Scale Insurance LLM Achieving Verifiable Domain Mastery and Hallucination Control without Competence Trade-offs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"INS-S1 reaches state-of-the-art insurance domain performance while preserving general capabilities and holding hallucinations to 0.6%.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jingjing Huo, Jun Li, Pan Liu, Qian Zhu, Wanqing Xu, Wenyan Yang, Xinnan Guo, Xuan Lin","submitted_at":"2026-03-15T16:13:37Z","abstract_excerpt":"Adapting Large Language Models (LLMs) to high-stakes vertical domains like insurance presents a significant challenge: scenarios demand strict adherence to complex regulations and business logic with zero tolerance for hallucinations. Existing approaches often suffer from a Competency Trade-off - sacrificing general intelligence for domain expertise - or rely heavily on RAG without intrinsic reasoning. To bridge this gap, we present INS-S1, an insurance-specific LLM family trained via a novel end-to-end alignment paradigm. Our approach features two methodological innovations: (1) A Verifiable "},"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":true},"canonical_record":{"source":{"id":"2603.14463","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-15T16:13:37Z","cross_cats_sorted":[],"title_canon_sha256":"3db4863f80ab015d67dab25ace0e8557b4bdb2a58c656a1829e8bfe92454a953","abstract_canon_sha256":"660096aafeba08ed6700c33a3eeb4c1f4b2c35e62294d7caa38b1d7e503b7b9f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T01:09:58.759062Z","signature_b64":"+CNcp1GpNWnDnh/wnGBMgRQh6LEXQSgQ/d+DnNF0mBK1WWBebgyhJ+wrsGEuojzuF12Q53osZNldY/I4qzt8DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7b9379f42f095071aca6a9f3e844cb9273771717d57bb0a5bd98440d3bbe9af1","last_reissued_at":"2026-06-10T01:09:58.758049Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T01:09:58.758049Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"An Industrial-Scale Insurance LLM Achieving Verifiable Domain Mastery and Hallucination Control without Competence Trade-offs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"INS-S1 reaches state-of-the-art insurance domain performance while preserving general capabilities and holding hallucinations to 0.6%.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jingjing Huo, Jun Li, Pan Liu, Qian Zhu, Wanqing Xu, Wenyan Yang, Xinnan Guo, Xuan Lin","submitted_at":"2026-03-15T16:13:37Z","abstract_excerpt":"Adapting Large Language Models (LLMs) to high-stakes vertical domains like insurance presents a significant challenge: scenarios demand strict adherence to complex regulations and business logic with zero tolerance for hallucinations. Existing approaches often suffer from a Competency Trade-off - sacrificing general intelligence for domain expertise - or rely heavily on RAG without intrinsic reasoning. To bridge this gap, we present INS-S1, an insurance-specific LLM family trained via a novel end-to-end alignment paradigm. Our approach features two methodological innovations: (1) A Verifiable "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"INS-S1 achieves SOTA performance on domain tasks, significantly outperforming DeepSeek-R1 and Gemini-2.5-Pro. Crucially, it maintains top-tier general capabilities and achieves a record-low 0.6% hallucination rate (HHEM).","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the Progressive SFT-RL Curriculum Framework with dynamic data annealing and the mix of RLVR and RLAIF can enforce domain constraints and prevent catastrophic forgetting without any hidden trade-offs or post-training degradation.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"INS-S1 is an insurance LLM that reaches SOTA domain performance with 0.6% hallucination rate and no loss in general intelligence via a new end-to-end alignment method and the INSEva benchmark.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"INS-S1 reaches state-of-the-art insurance domain performance while preserving general capabilities and holding hallucinations to 0.6%.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e035fb9b010ec9cddfc28cfe40c8aa4e325fb951acd08d9686172bb6ac8c8836"},"source":{"id":"2603.14463","kind":"arxiv","version":2},"verdict":{"id":"aabb3ae5-d2f3-4207-b879-51d468f21998","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T11:14:51.326694Z","strongest_claim":"INS-S1 achieves SOTA performance on domain tasks, significantly outperforming DeepSeek-R1 and Gemini-2.5-Pro. Crucially, it maintains top-tier general capabilities and achieves a record-low 0.6% hallucination rate (HHEM).","one_line_summary":"INS-S1 is an insurance LLM that reaches SOTA domain performance with 0.6% hallucination rate and no loss in general intelligence via a new end-to-end alignment method and the INSEva benchmark.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the Progressive SFT-RL Curriculum Framework with dynamic data annealing and the mix of RLVR and RLAIF can enforce domain constraints and prevent catastrophic forgetting without any hidden trade-offs or post-training degradation.","pith_extraction_headline":"INS-S1 reaches state-of-the-art insurance domain performance while preserving general capabilities and holding hallucinations to 0.6%."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2603.14463/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":2,"snapshot_sha256":"edb81a9fb4d4d5996c27c0aac04b1bae512ec173648191da069d644dcd19f58d"},"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":"2603.14463","created_at":"2026-06-10T01:09:58.758171+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.14463v2","created_at":"2026-06-10T01:09:58.758171+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.14463","created_at":"2026-06-10T01:09:58.758171+00:00"},{"alias_kind":"pith_short_12","alias_value":"POJXT5BPBFIH","created_at":"2026-06-10T01:09:58.758171+00:00"},{"alias_kind":"pith_short_16","alias_value":"POJXT5BPBFIHDLFG","created_at":"2026-06-10T01:09:58.758171+00:00"},{"alias_kind":"pith_short_8","alias_value":"POJXT5BP","created_at":"2026-06-10T01:09:58.758171+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/POJXT5BPBFIHDLFGVHZ6QRGLSJ","json":"https://pith.science/pith/POJXT5BPBFIHDLFGVHZ6QRGLSJ.json","graph_json":"https://pith.science/api/pith-number/POJXT5BPBFIHDLFGVHZ6QRGLSJ/graph.json","events_json":"https://pith.science/api/pith-number/POJXT5BPBFIHDLFGVHZ6QRGLSJ/events.json","paper":"https://pith.science/paper/POJXT5BP"},"agent_actions":{"view_html":"https://pith.science/pith/POJXT5BPBFIHDLFGVHZ6QRGLSJ","download_json":"https://pith.science/pith/POJXT5BPBFIHDLFGVHZ6QRGLSJ.json","view_paper":"https://pith.science/paper/POJXT5BP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.14463&json=true","fetch_graph":"https://pith.science/api/pith-number/POJXT5BPBFIHDLFGVHZ6QRGLSJ/graph.json","fetch_events":"https://pith.science/api/pith-number/POJXT5BPBFIHDLFGVHZ6QRGLSJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/POJXT5BPBFIHDLFGVHZ6QRGLSJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/POJXT5BPBFIHDLFGVHZ6QRGLSJ/action/storage_attestation","attest_author":"https://pith.science/pith/POJXT5BPBFIHDLFGVHZ6QRGLSJ/action/author_attestation","sign_citation":"https://pith.science/pith/POJXT5BPBFIHDLFGVHZ6QRGLSJ/action/citation_signature","submit_replication":"https://pith.science/pith/POJXT5BPBFIHDLFGVHZ6QRGLSJ/action/replication_record"}},"created_at":"2026-06-10T01:09:58.758171+00:00","updated_at":"2026-06-10T01:09:58.758171+00:00"}