{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:QCOUYGNIS7JKH43SRCPN3FJXQX","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":"5e1f223efcb3edf967bd36756bf2552c45f70304a93a49e93e195c4a894cab3a","cross_cats_sorted":["cs.CL","cs.CR","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2023-05-27T11:00:15Z","title_canon_sha256":"ce5e70c66e2e53e1bd2cab708309d0530badb51ea85040c0561e8983d987ff6c"},"schema_version":"1.0","source":{"id":"2305.17444","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.17444","created_at":"2026-07-05T06:14:44Z"},{"alias_kind":"arxiv_version","alias_value":"2305.17444v1","created_at":"2026-07-05T06:14:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.17444","created_at":"2026-07-05T06:14:44Z"},{"alias_kind":"pith_short_12","alias_value":"QCOUYGNIS7JK","created_at":"2026-07-05T06:14:44Z"},{"alias_kind":"pith_short_16","alias_value":"QCOUYGNIS7JKH43S","created_at":"2026-07-05T06:14:44Z"},{"alias_kind":"pith_short_8","alias_value":"QCOUYGNI","created_at":"2026-07-05T06:14:44Z"}],"graph_snapshots":[{"event_id":"sha256:149bd5abb99fe079e4f3edf8342569794169ec4f9b97f8bcbee447fe07e42b5c","target":"graph","created_at":"2026-07-05T06:14:44Z","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/2305.17444/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The deployment of large-scale generative models is often restricted by their potential risk of causing harm to users in unpredictable ways. We focus on the problem of black-box red teaming, where a red team generates test cases and interacts with the victim model to discover a diverse set of failures with limited query access. Existing red teaming methods construct test cases based on human supervision or language model (LM) and query all test cases in a brute-force manner without incorporating any information from past evaluations, resulting in a prohibitively large number of queries. To this","authors_text":"Deokjae Lee, Hwaran Lee, Hyun Oh Song, Jin-Hwa Kim, Jung-Woo Ha, JunYeong Lee, Sang-Woo Lee","cross_cats":["cs.CL","cs.CR","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2023-05-27T11:00:15Z","title":"Query-Efficient Black-Box Red Teaming via Bayesian Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.17444","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":"sha256:78b6607f4dbb9ca38574ab13e7c293bd94a30c050bb6908e6e8b1492d1de978d","target":"record","created_at":"2026-07-05T06:14:44Z","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":"5e1f223efcb3edf967bd36756bf2552c45f70304a93a49e93e195c4a894cab3a","cross_cats_sorted":["cs.CL","cs.CR","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2023-05-27T11:00:15Z","title_canon_sha256":"ce5e70c66e2e53e1bd2cab708309d0530badb51ea85040c0561e8983d987ff6c"},"schema_version":"1.0","source":{"id":"2305.17444","kind":"arxiv","version":1}},"canonical_sha256":"809d4c19a897d2a3f372889edd953785e7847c58c1aad67175c4656e4ffe29f7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"809d4c19a897d2a3f372889edd953785e7847c58c1aad67175c4656e4ffe29f7","first_computed_at":"2026-07-05T06:14:44.550611Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:14:44.550611Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"AnK3XHaLy4fu1jo5wtEzAd6Kst/j81htzSYVeeK6iTTrmxU9CKfQn6hKvuD5+zvbDSkVdNLDWcYUGPhOkukuBA==","signature_status":"signed_v1","signed_at":"2026-07-05T06:14:44.551060Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.17444","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:78b6607f4dbb9ca38574ab13e7c293bd94a30c050bb6908e6e8b1492d1de978d","sha256:149bd5abb99fe079e4f3edf8342569794169ec4f9b97f8bcbee447fe07e42b5c"],"state_sha256":"2aec31a40c4957351be9680d6a3344e27376978367d69969861296fff93bc7b7"}