{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VTA4WQTAHLGOEX7OVNORZEHB7G","short_pith_number":"pith:VTA4WQTA","schema_version":"1.0","canonical_sha256":"acc1cb42603acce25feeab5d1c90e1f984f1dbcdd4c2159b0384b5bae6b28634","source":{"kind":"arxiv","id":"2508.18384","version":1},"attestation_state":"computed","paper":{"title":"Backprompting: Leveraging Synthetic Production Data for Health Advice Guardrails","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Anna Lisa Gentile, Chad DeLuca, Guang-Jie Ren, Kellen Tan Cheng","submitted_at":"2025-08-25T18:17:00Z","abstract_excerpt":"The pervasiveness of large language models (LLMs) in enterprise settings has also brought forth a significant amount of risks associated with their usage. Guardrails technologies aim to mitigate this risk by filtering LLMs' input/output text through various detectors. However, developing and maintaining robust detectors faces many challenges, one of which is the difficulty in acquiring production-quality labeled data on real LLM outputs prior to deployment. In this work, we propose backprompting, a simple yet intuitive solution to generate production-like labeled data for health advice guardra"},"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":"2508.18384","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-25T18:17:00Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d29ffa1def0c4d45a5a66343001d559f2bdf4f12e89708549b1519efe2f90562","abstract_canon_sha256":"5698bb4217c504b2cea5a6f2fb3020e74491e5ee94338ea46bcda833911752ea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:59:23.324997Z","signature_b64":"QZjaFGafdF6IymKJ29P/GQ00l1VRc5JO4tgeMpqv8ZPw46eoS6LV0646Pf1MwXmpVUti+nzXA5KgzCGbzAWGDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"acc1cb42603acce25feeab5d1c90e1f984f1dbcdd4c2159b0384b5bae6b28634","last_reissued_at":"2026-07-05T11:59:23.324531Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:59:23.324531Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Backprompting: Leveraging Synthetic Production Data for Health Advice Guardrails","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Anna Lisa Gentile, Chad DeLuca, Guang-Jie Ren, Kellen Tan Cheng","submitted_at":"2025-08-25T18:17:00Z","abstract_excerpt":"The pervasiveness of large language models (LLMs) in enterprise settings has also brought forth a significant amount of risks associated with their usage. Guardrails technologies aim to mitigate this risk by filtering LLMs' input/output text through various detectors. However, developing and maintaining robust detectors faces many challenges, one of which is the difficulty in acquiring production-quality labeled data on real LLM outputs prior to deployment. In this work, we propose backprompting, a simple yet intuitive solution to generate production-like labeled data for health advice guardra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.18384","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/2508.18384/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":"2508.18384","created_at":"2026-07-05T11:59:23.324595+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.18384v1","created_at":"2026-07-05T11:59:23.324595+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.18384","created_at":"2026-07-05T11:59:23.324595+00:00"},{"alias_kind":"pith_short_12","alias_value":"VTA4WQTAHLGO","created_at":"2026-07-05T11:59:23.324595+00:00"},{"alias_kind":"pith_short_16","alias_value":"VTA4WQTAHLGOEX7O","created_at":"2026-07-05T11:59:23.324595+00:00"},{"alias_kind":"pith_short_8","alias_value":"VTA4WQTA","created_at":"2026-07-05T11:59:23.324595+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VTA4WQTAHLGOEX7OVNORZEHB7G","json":"https://pith.science/pith/VTA4WQTAHLGOEX7OVNORZEHB7G.json","graph_json":"https://pith.science/api/pith-number/VTA4WQTAHLGOEX7OVNORZEHB7G/graph.json","events_json":"https://pith.science/api/pith-number/VTA4WQTAHLGOEX7OVNORZEHB7G/events.json","paper":"https://pith.science/paper/VTA4WQTA"},"agent_actions":{"view_html":"https://pith.science/pith/VTA4WQTAHLGOEX7OVNORZEHB7G","download_json":"https://pith.science/pith/VTA4WQTAHLGOEX7OVNORZEHB7G.json","view_paper":"https://pith.science/paper/VTA4WQTA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.18384&json=true","fetch_graph":"https://pith.science/api/pith-number/VTA4WQTAHLGOEX7OVNORZEHB7G/graph.json","fetch_events":"https://pith.science/api/pith-number/VTA4WQTAHLGOEX7OVNORZEHB7G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VTA4WQTAHLGOEX7OVNORZEHB7G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VTA4WQTAHLGOEX7OVNORZEHB7G/action/storage_attestation","attest_author":"https://pith.science/pith/VTA4WQTAHLGOEX7OVNORZEHB7G/action/author_attestation","sign_citation":"https://pith.science/pith/VTA4WQTAHLGOEX7OVNORZEHB7G/action/citation_signature","submit_replication":"https://pith.science/pith/VTA4WQTAHLGOEX7OVNORZEHB7G/action/replication_record"}},"created_at":"2026-07-05T11:59:23.324595+00:00","updated_at":"2026-07-05T11:59:23.324595+00:00"}