{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:FNA6XVC5675OQ3HXT4SJ363LM2","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":"9d35d37da79dc781c2abfae62ea3b384f7a57ab2f6ea13539a9d3c42fbda181e","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-13T15:56:50Z","title_canon_sha256":"0f5e4000ce009452d6ad60460b1fadd8ba6441bb5ad640e9e47296914bd79b15"},"schema_version":"1.0","source":{"id":"2309.07045","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.07045","created_at":"2026-07-05T08:35:36Z"},{"alias_kind":"arxiv_version","alias_value":"2309.07045v2","created_at":"2026-07-05T08:35:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.07045","created_at":"2026-07-05T08:35:36Z"},{"alias_kind":"pith_short_12","alias_value":"FNA6XVC5675O","created_at":"2026-07-05T08:35:36Z"},{"alias_kind":"pith_short_16","alias_value":"FNA6XVC5675OQ3HX","created_at":"2026-07-05T08:35:36Z"},{"alias_kind":"pith_short_8","alias_value":"FNA6XVC5","created_at":"2026-07-05T08:35:36Z"}],"graph_snapshots":[{"event_id":"sha256:621a7f3d8e65cd748dfc988df318dd6cf61ab848b1fd9114015781232ca35a0b","target":"graph","created_at":"2026-07-05T08:35:36Z","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/2309.07045/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the rapid development of Large Language Models (LLMs), increasing attention has been paid to their safety concerns. Consequently, evaluating the safety of LLMs has become an essential task for facilitating the broad applications of LLMs. Nevertheless, the absence of comprehensive safety evaluation benchmarks poses a significant impediment to effectively assess and enhance the safety of LLMs. In this work, we present SafetyBench, a comprehensive benchmark for evaluating the safety of LLMs, which comprises 11,435 diverse multiple choice questions spanning across 7 distinct categories of saf","authors_text":"Chong Long, Jie Tang, Leqi Lei, Lindong Wu, Minlie Huang, Rui Sun, Xiao Liu, Xuanyu Lei, Yongkang Huang, Zhexin Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-13T15:56:50Z","title":"SafetyBench: Evaluating the Safety of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.07045","kind":"arxiv","version":2},"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:f4494bb1f230e1527f36b6121a3620ca48648189a7e7b6673abea0054cd7bdcb","target":"record","created_at":"2026-07-05T08:35:36Z","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":"9d35d37da79dc781c2abfae62ea3b384f7a57ab2f6ea13539a9d3c42fbda181e","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-13T15:56:50Z","title_canon_sha256":"0f5e4000ce009452d6ad60460b1fadd8ba6441bb5ad640e9e47296914bd79b15"},"schema_version":"1.0","source":{"id":"2309.07045","kind":"arxiv","version":2}},"canonical_sha256":"2b41ebd45df7fae86cf79f249dfb6b6690ca87c8d1e246658f8b822058d29dca","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2b41ebd45df7fae86cf79f249dfb6b6690ca87c8d1e246658f8b822058d29dca","first_computed_at":"2026-07-05T08:35:36.835986Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:35:36.835986Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vWo6soRh6tGiVvFG2I8Y++r7WwSekFtN6Yu65ylcOu5xjovisSzo9DOoGgI/fNqK+Ql21ixT32OZFIxiMaOaCA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:35:36.836470Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.07045","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f4494bb1f230e1527f36b6121a3620ca48648189a7e7b6673abea0054cd7bdcb","sha256:621a7f3d8e65cd748dfc988df318dd6cf61ab848b1fd9114015781232ca35a0b"],"state_sha256":"d3481225dc85ff0746246d80ccba0561bbcd4bc8cdef32279dd1ed7f683eae75"}