{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:COFYDDMXGPWBKT4S6YFB4GVEKA","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":"4ae45f13bfc8df7779e26b5cfdcf9daaa20d314e0233ec125400969113b9ad7e","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-24T12:19:40Z","title_canon_sha256":"e0d5c0a3a5e46a9c3d32b32c927a4efb8f2c852958e2957ec807eba80ab70da0"},"schema_version":"1.0","source":{"id":"2606.25750","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.25750","created_at":"2026-06-25T01:18:14Z"},{"alias_kind":"arxiv_version","alias_value":"2606.25750v1","created_at":"2026-06-25T01:18:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.25750","created_at":"2026-06-25T01:18:14Z"},{"alias_kind":"pith_short_12","alias_value":"COFYDDMXGPWB","created_at":"2026-06-25T01:18:14Z"},{"alias_kind":"pith_short_16","alias_value":"COFYDDMXGPWBKT4S","created_at":"2026-06-25T01:18:14Z"},{"alias_kind":"pith_short_8","alias_value":"COFYDDMX","created_at":"2026-06-25T01:18:14Z"}],"graph_snapshots":[{"event_id":"sha256:45402d3e3207cb42899a4bfd799300ffb927b018a6972d557aa67cb8899e7803","target":"graph","created_at":"2026-06-25T01:18:14Z","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/2606.25750/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Safety evaluation of large language models (LLMs) is commonly performed by querying models with unsafe or jailbreak prompts and judging whether their outputs violate a safety policy. Although useful, output-level evaluation is expensive, sensitive to judge choice, and easily tied to fixed question banks. We propose **SafeVec**, a white-box evaluation procedure that measures safety from internal representations rather than generated answers. **SafeVec** first extracts layer-wise refusal directions from a safety-aligned reference model, then selects stable layer windows where safe and unsafe beh","authors_text":"Chang-Chieh Huang, Chia-Mu Yu, Wei-Bin Lee, Yan-Lun Chen","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-24T12:19:40Z","title":"RAS: Measuring LLM Safety Through Refusal Alignment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.25750","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:b41345b1ac438cc8b9fb2c3c6dbc6fb89ec977a00afb32236f7f2f1acf346124","target":"record","created_at":"2026-06-25T01:18:14Z","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":"4ae45f13bfc8df7779e26b5cfdcf9daaa20d314e0233ec125400969113b9ad7e","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-06-24T12:19:40Z","title_canon_sha256":"e0d5c0a3a5e46a9c3d32b32c927a4efb8f2c852958e2957ec807eba80ab70da0"},"schema_version":"1.0","source":{"id":"2606.25750","kind":"arxiv","version":1}},"canonical_sha256":"138b818d9733ec154f92f60a1e1aa450091211287af18650d843bf4f6eac9155","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"138b818d9733ec154f92f60a1e1aa450091211287af18650d843bf4f6eac9155","first_computed_at":"2026-06-25T01:18:14.270081Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-25T01:18:14.270081Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ufUXpDuO4rj+dGFQwAAmNGPp4ZTSJVYxZbIN/H6G44CQ6ylvA3rx69abkZ8Uhuz5A9VWkJlUC6BrnZ2YcQTbDg==","signature_status":"signed_v1","signed_at":"2026-06-25T01:18:14.270464Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.25750","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b41345b1ac438cc8b9fb2c3c6dbc6fb89ec977a00afb32236f7f2f1acf346124","sha256:45402d3e3207cb42899a4bfd799300ffb927b018a6972d557aa67cb8899e7803"],"state_sha256":"b8177e89dca710cbf88f86fc86c82c4c9c69b1c5e9767d1335a1983314ae089b"}