{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FSOPWZADYFN3S7ND6DR3YVVHXT","short_pith_number":"pith:FSOPWZAD","schema_version":"1.0","canonical_sha256":"2c9cfb6403c15bb97da3f0e3bc56a7bcf3edb3277214d3f76d847014e801afbc","source":{"kind":"arxiv","id":"2607.02079","version":1},"attestation_state":"computed","paper":{"title":"HaloGuard 1.0: An Open Weights Constitutional Classifier for Multilingual AI Safety","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ashmiya Lenin, Navaneeth Sangameswaran, Preetham S","submitted_at":"2026-07-02T12:21:16Z","abstract_excerpt":"We present HaloGuard 1.0, an open-weights implementation of the constitutional-classifier paradigm for input safety. It achieves state-of-the-art performance on English and multilingual prompt-safety benchmarks at roughly one-tenth the model size of current leading open guard models. The safety constitution is the organising structure of the corpus: a natural-language constitution of 46 policies and 2,940 subcategories drives synthetic data generation, with exhaustive one-to-one paired counterfactuals that hold topic and vocabulary fixed while flipping intent, a two-tier harmless design that s"},"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":"2607.02079","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-02T12:21:16Z","cross_cats_sorted":["cs.CR","cs.LG"],"title_canon_sha256":"616f38ee9df16611f11f0ca89d3c4aecccb3b5808ed4fc9506d4ef0d0d8e7fb2","abstract_canon_sha256":"939477974d0884c369e79c05bf79c017002dbde4c00537411e7f4b688cce3435"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:41.594303Z","signature_b64":"/ZeBq/W40VENcNX5FOHYw7f5Y1Q7oSy0CgM9tEtYug275Wh1mlVeMijisxKQ2xdxLQ9NQA1a7OBpyhvgh5uZAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2c9cfb6403c15bb97da3f0e3bc56a7bcf3edb3277214d3f76d847014e801afbc","last_reissued_at":"2026-07-03T01:17:41.593873Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:41.593873Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HaloGuard 1.0: An Open Weights Constitutional Classifier for Multilingual AI Safety","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ashmiya Lenin, Navaneeth Sangameswaran, Preetham S","submitted_at":"2026-07-02T12:21:16Z","abstract_excerpt":"We present HaloGuard 1.0, an open-weights implementation of the constitutional-classifier paradigm for input safety. It achieves state-of-the-art performance on English and multilingual prompt-safety benchmarks at roughly one-tenth the model size of current leading open guard models. The safety constitution is the organising structure of the corpus: a natural-language constitution of 46 policies and 2,940 subcategories drives synthetic data generation, with exhaustive one-to-one paired counterfactuals that hold topic and vocabulary fixed while flipping intent, a two-tier harmless design that s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02079","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/2607.02079/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":"2607.02079","created_at":"2026-07-03T01:17:41.593935+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.02079v1","created_at":"2026-07-03T01:17:41.593935+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02079","created_at":"2026-07-03T01:17:41.593935+00:00"},{"alias_kind":"pith_short_12","alias_value":"FSOPWZADYFN3","created_at":"2026-07-03T01:17:41.593935+00:00"},{"alias_kind":"pith_short_16","alias_value":"FSOPWZADYFN3S7ND","created_at":"2026-07-03T01:17:41.593935+00:00"},{"alias_kind":"pith_short_8","alias_value":"FSOPWZAD","created_at":"2026-07-03T01:17:41.593935+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/FSOPWZADYFN3S7ND6DR3YVVHXT","json":"https://pith.science/pith/FSOPWZADYFN3S7ND6DR3YVVHXT.json","graph_json":"https://pith.science/api/pith-number/FSOPWZADYFN3S7ND6DR3YVVHXT/graph.json","events_json":"https://pith.science/api/pith-number/FSOPWZADYFN3S7ND6DR3YVVHXT/events.json","paper":"https://pith.science/paper/FSOPWZAD"},"agent_actions":{"view_html":"https://pith.science/pith/FSOPWZADYFN3S7ND6DR3YVVHXT","download_json":"https://pith.science/pith/FSOPWZADYFN3S7ND6DR3YVVHXT.json","view_paper":"https://pith.science/paper/FSOPWZAD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.02079&json=true","fetch_graph":"https://pith.science/api/pith-number/FSOPWZADYFN3S7ND6DR3YVVHXT/graph.json","fetch_events":"https://pith.science/api/pith-number/FSOPWZADYFN3S7ND6DR3YVVHXT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FSOPWZADYFN3S7ND6DR3YVVHXT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FSOPWZADYFN3S7ND6DR3YVVHXT/action/storage_attestation","attest_author":"https://pith.science/pith/FSOPWZADYFN3S7ND6DR3YVVHXT/action/author_attestation","sign_citation":"https://pith.science/pith/FSOPWZADYFN3S7ND6DR3YVVHXT/action/citation_signature","submit_replication":"https://pith.science/pith/FSOPWZADYFN3S7ND6DR3YVVHXT/action/replication_record"}},"created_at":"2026-07-03T01:17:41.593935+00:00","updated_at":"2026-07-03T01:17:41.593935+00:00"}