{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ULCZAJ3TLXMQ7VGDP466XJG7UV","short_pith_number":"pith:ULCZAJ3T","schema_version":"1.0","canonical_sha256":"a2c59027735dd90fd4c37f3deba4dfa561fb46d0467720402ef8184bfa663fe7","source":{"kind":"arxiv","id":"2508.09893","version":1},"attestation_state":"computed","paper":{"title":"RAGulating Compliance: A Multi-Agent Knowledge Graph for Regulatory QA","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bhavik Agarwal, Hemant Sunil Jomraj, Jack Krolick, Simone Kaplunov, Viktoria Rojkova","submitted_at":"2025-08-13T15:51:05Z","abstract_excerpt":"Regulatory compliance question answering (QA) requires precise, verifiable information, and domain-specific expertise, posing challenges for Large Language Models (LLMs). In this work, we present a novel multi-agent framework that integrates a Knowledge Graph (KG) of Regulatory triplets with Retrieval-Augmented Generation (RAG) to address these demands. First, agents build and maintain an ontology-free KG by extracting subject--predicate--object (SPO) triplets from regulatory documents and systematically cleaning, normalizing, deduplicating, and updating them. Second, these triplets are embedd"},"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":"2508.09893","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-13T15:51:05Z","cross_cats_sorted":[],"title_canon_sha256":"949c950374290ae2a68250446bcc03b4427fa37480e16dfc35403819345fbe26","abstract_canon_sha256":"0b91f456ced9d88d61b30a8be6aa6e021517845c89a4d9edb8233a4d8c0e7ce0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:23.862417Z","signature_b64":"zJZH8YHMj8jFe8y6Ywi3p34Bl44+KVxT3d4jg5WKSdS+3Tqz1ChFnippo9k//EslnpC9BBEMHMkPb7LY8ZjJAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a2c59027735dd90fd4c37f3deba4dfa561fb46d0467720402ef8184bfa663fe7","last_reissued_at":"2026-07-05T11:53:23.861889Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:23.861889Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RAGulating Compliance: A Multi-Agent Knowledge Graph for Regulatory QA","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bhavik Agarwal, Hemant Sunil Jomraj, Jack Krolick, Simone Kaplunov, Viktoria Rojkova","submitted_at":"2025-08-13T15:51:05Z","abstract_excerpt":"Regulatory compliance question answering (QA) requires precise, verifiable information, and domain-specific expertise, posing challenges for Large Language Models (LLMs). In this work, we present a novel multi-agent framework that integrates a Knowledge Graph (KG) of Regulatory triplets with Retrieval-Augmented Generation (RAG) to address these demands. First, agents build and maintain an ontology-free KG by extracting subject--predicate--object (SPO) triplets from regulatory documents and systematically cleaning, normalizing, deduplicating, and updating them. Second, these triplets are embedd"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.09893","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/2508.09893/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":"2508.09893","created_at":"2026-07-05T11:53:23.861956+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.09893v1","created_at":"2026-07-05T11:53:23.861956+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.09893","created_at":"2026-07-05T11:53:23.861956+00:00"},{"alias_kind":"pith_short_12","alias_value":"ULCZAJ3TLXMQ","created_at":"2026-07-05T11:53:23.861956+00:00"},{"alias_kind":"pith_short_16","alias_value":"ULCZAJ3TLXMQ7VGD","created_at":"2026-07-05T11:53:23.861956+00:00"},{"alias_kind":"pith_short_8","alias_value":"ULCZAJ3T","created_at":"2026-07-05T11:53:23.861956+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08292","citing_title":"Reverse Engineering Compliance: A Dual-Graph Verification Framework for Auditing Legacy IT Security Concepts","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2606.03031","citing_title":"AUDITFLOW: Executable Symbolic Environments for Structured Financial Reporting Verification","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27713","citing_title":"Knowledge Graph Representations for LLM-Based Policy Compliance Reasoning","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ULCZAJ3TLXMQ7VGDP466XJG7UV","json":"https://pith.science/pith/ULCZAJ3TLXMQ7VGDP466XJG7UV.json","graph_json":"https://pith.science/api/pith-number/ULCZAJ3TLXMQ7VGDP466XJG7UV/graph.json","events_json":"https://pith.science/api/pith-number/ULCZAJ3TLXMQ7VGDP466XJG7UV/events.json","paper":"https://pith.science/paper/ULCZAJ3T"},"agent_actions":{"view_html":"https://pith.science/pith/ULCZAJ3TLXMQ7VGDP466XJG7UV","download_json":"https://pith.science/pith/ULCZAJ3TLXMQ7VGDP466XJG7UV.json","view_paper":"https://pith.science/paper/ULCZAJ3T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.09893&json=true","fetch_graph":"https://pith.science/api/pith-number/ULCZAJ3TLXMQ7VGDP466XJG7UV/graph.json","fetch_events":"https://pith.science/api/pith-number/ULCZAJ3TLXMQ7VGDP466XJG7UV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ULCZAJ3TLXMQ7VGDP466XJG7UV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ULCZAJ3TLXMQ7VGDP466XJG7UV/action/storage_attestation","attest_author":"https://pith.science/pith/ULCZAJ3TLXMQ7VGDP466XJG7UV/action/author_attestation","sign_citation":"https://pith.science/pith/ULCZAJ3TLXMQ7VGDP466XJG7UV/action/citation_signature","submit_replication":"https://pith.science/pith/ULCZAJ3TLXMQ7VGDP466XJG7UV/action/replication_record"}},"created_at":"2026-07-05T11:53:23.861956+00:00","updated_at":"2026-07-05T11:53:23.861956+00:00"}