{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:POJXT5BPBFIHDLFGVHZ6QRGLSJ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"660096aafeba08ed6700c33a3eeb4c1f4b2c35e62294d7caa38b1d7e503b7b9f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-15T16:13:37Z","title_canon_sha256":"3db4863f80ab015d67dab25ace0e8557b4bdb2a58c656a1829e8bfe92454a953"},"schema_version":"1.0","source":{"id":"2603.14463","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.14463","created_at":"2026-06-10T01:09:58Z"},{"alias_kind":"arxiv_version","alias_value":"2603.14463v2","created_at":"2026-06-10T01:09:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.14463","created_at":"2026-06-10T01:09:58Z"},{"alias_kind":"pith_short_12","alias_value":"POJXT5BPBFIH","created_at":"2026-06-10T01:09:58Z"},{"alias_kind":"pith_short_16","alias_value":"POJXT5BPBFIHDLFG","created_at":"2026-06-10T01:09:58Z"},{"alias_kind":"pith_short_8","alias_value":"POJXT5BP","created_at":"2026-06-10T01:09:58Z"}],"graph_snapshots":[{"event_id":"sha256:f1a9b81acf28c8334a5f2d106cef7c36d240068c5e3c948ecedaa3272026f676","target":"graph","created_at":"2026-06-10T01:09:58Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","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)."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"INS-S1 reaches state-of-the-art insurance domain performance while preserving general capabilities and holding hallucinations to 0.6%."}],"snapshot_sha256":"e035fb9b010ec9cddfc28cfe40c8aa4e325fb951acd08d9686172bb6ac8c8836"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"edb81a9fb4d4d5996c27c0aac04b1bae512ec173648191da069d644dcd19f58d"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2603.14463/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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 ","authors_text":"Jingjing Huo, Jun Li, Pan Liu, Qian Zhu, Wanqing Xu, Wenyan Yang, Xinnan Guo, Xuan Lin","cross_cats":[],"headline":"INS-S1 reaches state-of-the-art insurance domain performance while preserving general capabilities and holding hallucinations to 0.6%.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-15T16:13:37Z","title":"An Industrial-Scale Insurance LLM Achieving Verifiable Domain Mastery and Hallucination Control without Competence Trade-offs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.14463","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-15T11:14:51.326694Z","id":"aabb3ae5-d2f3-4207-b879-51d468f21998","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"INS-S1 reaches state-of-the-art insurance domain performance while preserving general capabilities and holding hallucinations to 0.6%.","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).","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."}},"verdict_id":"aabb3ae5-d2f3-4207-b879-51d468f21998"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8a68f3af18c84b07d45b7e8e9efb2f3d400a9ff93f6ccec1e65bf5ce8d189add","target":"record","created_at":"2026-06-10T01:09:58Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"660096aafeba08ed6700c33a3eeb4c1f4b2c35e62294d7caa38b1d7e503b7b9f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-15T16:13:37Z","title_canon_sha256":"3db4863f80ab015d67dab25ace0e8557b4bdb2a58c656a1829e8bfe92454a953"},"schema_version":"1.0","source":{"id":"2603.14463","kind":"arxiv","version":2}},"canonical_sha256":"7b9379f42f095071aca6a9f3e844cb9273771717d57bb0a5bd98440d3bbe9af1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7b9379f42f095071aca6a9f3e844cb9273771717d57bb0a5bd98440d3bbe9af1","first_computed_at":"2026-06-10T01:09:58.758049Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-10T01:09:58.758049Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+CNcp1GpNWnDnh/wnGBMgRQh6LEXQSgQ/d+DnNF0mBK1WWBebgyhJ+wrsGEuojzuF12Q53osZNldY/I4qzt8DA==","signature_status":"signed_v1","signed_at":"2026-06-10T01:09:58.759062Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.14463","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8a68f3af18c84b07d45b7e8e9efb2f3d400a9ff93f6ccec1e65bf5ce8d189add","sha256:f1a9b81acf28c8334a5f2d106cef7c36d240068c5e3c948ecedaa3272026f676"],"state_sha256":"fc471e392e230afe75b8a9681faa7f09a58ce7c1a1857f7cf4097a86b7cfd204"}