{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:4QSCSFX7AIPS47T4SUFWRUZO77","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":"22b7d9fe720352fd593f7b94ded9bf3c546f13111da7af86684b35ec2687ee46","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-02T16:23:44Z","title_canon_sha256":"219495cd39bd22e80bcdbc062a757afa04bf7b6fbbe9657a3711699de4c28e6e"},"schema_version":"1.0","source":{"id":"2403.01251","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.01251","created_at":"2026-07-05T09:32:42Z"},{"alias_kind":"arxiv_version","alias_value":"2403.01251v3","created_at":"2026-07-05T09:32:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.01251","created_at":"2026-07-05T09:32:42Z"},{"alias_kind":"pith_short_12","alias_value":"4QSCSFX7AIPS","created_at":"2026-07-05T09:32:42Z"},{"alias_kind":"pith_short_16","alias_value":"4QSCSFX7AIPS47T4","created_at":"2026-07-05T09:32:42Z"},{"alias_kind":"pith_short_8","alias_value":"4QSCSFX7","created_at":"2026-07-05T09:32:42Z"}],"graph_snapshots":[{"event_id":"sha256:bb20cedc6488fe5a466414e5267919bda78905880519fe1f73d885b0c8324de9","target":"graph","created_at":"2026-07-05T09:32:42Z","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/2403.01251/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Safety of Large Language Models (LLMs) has become a critical issue given their rapid progresses. Greedy Coordinate Gradient (GCG) is shown to be effective in constructing adversarial prompts to break the aligned LLMs, but optimization of GCG is time-consuming. To reduce the time cost of GCG and enable more comprehensive studies of LLM safety, in this work, we study a new algorithm called $\\texttt{Probe sampling}$. At the core of the algorithm is a mechanism that dynamically determines how similar a smaller draft model's predictions are to the target model's predictions for prompt candidates. W","authors_text":"Anirudh Goyal, Kenji Kawaguchi, Michael Shieh, Tianle Cai, Wenyue Zheng, Xuan Long Do, Yiran Zhao","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-02T16:23:44Z","title":"Accelerating Greedy Coordinate Gradient and General Prompt Optimization via Probe Sampling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.01251","kind":"arxiv","version":3},"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":"sha256:170124238bf9a0a44e96d03801407b87fc244f54a4782b4fe3e885ba576fbd72","target":"record","created_at":"2026-07-05T09:32:42Z","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":"22b7d9fe720352fd593f7b94ded9bf3c546f13111da7af86684b35ec2687ee46","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-02T16:23:44Z","title_canon_sha256":"219495cd39bd22e80bcdbc062a757afa04bf7b6fbbe9657a3711699de4c28e6e"},"schema_version":"1.0","source":{"id":"2403.01251","kind":"arxiv","version":3}},"canonical_sha256":"e4242916ff021f2e7e7c950b68d32efff50f4c1d761768e9c33c32df22f75f61","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e4242916ff021f2e7e7c950b68d32efff50f4c1d761768e9c33c32df22f75f61","first_computed_at":"2026-07-05T09:32:42.500323Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:32:42.500323Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"BcwM/oqqMSF2HV0sCd6Hf7vcAZONleOCQncJfyeaEfKExXQPeUJdwNtxhA2EbjSqKYHQXPQCRstXC6A1Vf8yAA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:32:42.501422Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.01251","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:170124238bf9a0a44e96d03801407b87fc244f54a4782b4fe3e885ba576fbd72","sha256:bb20cedc6488fe5a466414e5267919bda78905880519fe1f73d885b0c8324de9"],"state_sha256":"5a46493ebe380cd82eeda60a5fd8997561f27c0b204ffef403b44cb6051943a1"}