{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:YNFBNSISMWIYBMDYY54LDWF7QZ","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":"bbf84ee410771af71fc99d65cfbaea4cf5b1c640eb50bb8428d51a2bd2a929c2","cross_cats_sorted":["cs.AI","cs.CR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-15T02:54:49Z","title_canon_sha256":"ddd08cbe4da43c72b304c9d8ccdda70931e8e3f90494edc448e125eb831616b9"},"schema_version":"1.0","source":{"id":"2402.09674","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.09674","created_at":"2026-07-05T07:45:34Z"},{"alias_kind":"arxiv_version","alias_value":"2402.09674v1","created_at":"2026-07-05T07:45:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.09674","created_at":"2026-07-05T07:45:34Z"},{"alias_kind":"pith_short_12","alias_value":"YNFBNSISMWIY","created_at":"2026-07-05T07:45:34Z"},{"alias_kind":"pith_short_16","alias_value":"YNFBNSISMWIYBMDY","created_at":"2026-07-05T07:45:34Z"},{"alias_kind":"pith_short_8","alias_value":"YNFBNSIS","created_at":"2026-07-05T07:45:34Z"}],"graph_snapshots":[{"event_id":"sha256:da621baa8692a395281bd8dcc2d7c515d94f6dff30070d1f242559dccdb8d034","target":"graph","created_at":"2026-07-05T07:45:34Z","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/2402.09674/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have surged in popularity in recent months, but they have demonstrated concerning capabilities to generate harmful content when manipulated. While techniques like safety fine-tuning aim to minimize harmful use, recent works have shown that LLMs remain vulnerable to attacks that elicit toxic responses. In this work, we introduce the Proxy-Guided Attack on LLMs (PAL), the first optimization-based attack on LLMs in a black-box query-only setting. In particular, it relies on a surrogate model to guide the optimization and a sophisticated loss designed for real-world LL","authors_text":"Alexandre Araujo, Chawin Sitawarin, David Wagner, Norman Mu","cross_cats":["cs.AI","cs.CR","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-15T02:54:49Z","title":"PAL: Proxy-Guided Black-Box Attack on Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.09674","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":"sha256:15e7ce5a1ff7e80d100e7bbc04dba0108c4d481da5c7bccdbfa16097a20ccea8","target":"record","created_at":"2026-07-05T07:45:34Z","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":"bbf84ee410771af71fc99d65cfbaea4cf5b1c640eb50bb8428d51a2bd2a929c2","cross_cats_sorted":["cs.AI","cs.CR","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-15T02:54:49Z","title_canon_sha256":"ddd08cbe4da43c72b304c9d8ccdda70931e8e3f90494edc448e125eb831616b9"},"schema_version":"1.0","source":{"id":"2402.09674","kind":"arxiv","version":1}},"canonical_sha256":"c34a16c912659180b078c778b1d8bf866383c0336c9c7bedb079c69e770d2e20","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c34a16c912659180b078c778b1d8bf866383c0336c9c7bedb079c69e770d2e20","first_computed_at":"2026-07-05T07:45:34.738053Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:45:34.738053Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yDWFBAFHC5PPmF2TCy/0uJouP0AjZ2+tAWBGYr5Tmzlncprmx7AitAQExta2Z2FTCL7w+iEvXDEGSxVhcexDAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:45:34.738473Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.09674","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:15e7ce5a1ff7e80d100e7bbc04dba0108c4d481da5c7bccdbfa16097a20ccea8","sha256:da621baa8692a395281bd8dcc2d7c515d94f6dff30070d1f242559dccdb8d034"],"state_sha256":"6fd58f1cd7a33e0e94b4741f400ed66817ec30c809815b69c2d7535db77062a5"}