{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:RFUG5JBNRA72UOLZ45EOZLODJ4","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":"2aa6e47fb1d1c661757f3e7b583512c07b176b139ed37a11e40c055bef7e64eb","cross_cats_sorted":["cs.AI","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-25T04:45:44Z","title_canon_sha256":"4404a1f05b8cdd4a70401f944b48c1c43f394076b25409cb141dd578658a54e7"},"schema_version":"1.0","source":{"id":"2605.25420","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.25420","created_at":"2026-05-26T02:04:34Z"},{"alias_kind":"arxiv_version","alias_value":"2605.25420v1","created_at":"2026-05-26T02:04:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.25420","created_at":"2026-05-26T02:04:34Z"},{"alias_kind":"pith_short_12","alias_value":"RFUG5JBNRA72","created_at":"2026-05-26T02:04:34Z"},{"alias_kind":"pith_short_16","alias_value":"RFUG5JBNRA72UOLZ","created_at":"2026-05-26T02:04:34Z"},{"alias_kind":"pith_short_8","alias_value":"RFUG5JBN","created_at":"2026-05-26T02:04:34Z"}],"graph_snapshots":[{"event_id":"sha256:f2945e755427270635df2f07b97aa4b1ac6cd4ab6666fb7f3aa523b3a33ebad8","target":"graph","created_at":"2026-05-26T02:04:34Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2605.25420/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language model safety evaluation remains heavily English-centered, leaving low-resource languages under-measured even when models are deployed globally. We evaluate four open-weight instruction-tuned models on SomaliBench v0, a native-author-verified benchmark of 100 harmful-intent prompts paired across English and Somali. Each of Llama-3.1-8B-Instruct, Gemma-2-9B-Instruct, Qwen-2.5-7B-Instruct, and Aya-23-8B is run locally with temperature 0 and the same English \"helpful, harmless, and honest\" (HHH) system prompt. A pinned Claude Sonnet snapshot (claude-sonnet-4-5-20250929) classifies e","authors_text":"Khalid Yusuf Dahir","cross_cats":["cs.AI","cs.CY"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-25T04:45:44Z","title":"SomaliBench Eval: Measuring English-to-Somali Refusal Gaps in Open-Weight Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.25420","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:fc6e6809155193427087c11ea0a2efeb7997410deb90545f0316f843d21c203a","target":"record","created_at":"2026-05-26T02:04:34Z","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":"2aa6e47fb1d1c661757f3e7b583512c07b176b139ed37a11e40c055bef7e64eb","cross_cats_sorted":["cs.AI","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-25T04:45:44Z","title_canon_sha256":"4404a1f05b8cdd4a70401f944b48c1c43f394076b25409cb141dd578658a54e7"},"schema_version":"1.0","source":{"id":"2605.25420","kind":"arxiv","version":1}},"canonical_sha256":"89686ea42d883faa3979e748ecadc34f1009e48dea79b97cad36298a0b21d8a9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"89686ea42d883faa3979e748ecadc34f1009e48dea79b97cad36298a0b21d8a9","first_computed_at":"2026-05-26T02:04:34.079569Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T02:04:34.079569Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"E7nvlaDqw45fbySZWKzSZqiqUTMehVeXw7lT5nvAQVFfL/zPUNwZ4sFEhK2B8GCmq3vdpJGNJ13UCOuK/ysGDQ==","signature_status":"signed_v1","signed_at":"2026-05-26T02:04:34.080543Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.25420","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fc6e6809155193427087c11ea0a2efeb7997410deb90545f0316f843d21c203a","sha256:f2945e755427270635df2f07b97aa4b1ac6cd4ab6666fb7f3aa523b3a33ebad8"],"state_sha256":"5e8b173af5a2a6fd092dc733f951ab213c650667d62c366864e57649d32459b3"}