{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:6FMPDCBXIFB3SNL4P63WUWJ5HY","short_pith_number":"pith:6FMPDCBX","canonical_record":{"source":{"id":"2402.09154","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-14T13:13:26Z","cross_cats_sorted":[],"title_canon_sha256":"7fe82e64af51279a4c2fd8c93b64da2f08b17729f9c64bbe6c4ef189e402c6e4","abstract_canon_sha256":"04647cc40c6ec86f14d4147814d65b55465ff8907f31b5965d9b3799842ffaf2"},"schema_version":"1.0"},"canonical_sha256":"f158f188374143b9357c7fb76a593d3e28875063ac105a30263b778314676cfc","source":{"kind":"arxiv","id":"2402.09154","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.09154","created_at":"2026-07-05T10:22:44Z"},{"alias_kind":"arxiv_version","alias_value":"2402.09154v2","created_at":"2026-07-05T10:22:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.09154","created_at":"2026-07-05T10:22:44Z"},{"alias_kind":"pith_short_12","alias_value":"6FMPDCBXIFB3","created_at":"2026-07-05T10:22:44Z"},{"alias_kind":"pith_short_16","alias_value":"6FMPDCBXIFB3SNL4","created_at":"2026-07-05T10:22:44Z"},{"alias_kind":"pith_short_8","alias_value":"6FMPDCBX","created_at":"2026-07-05T10:22:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:6FMPDCBXIFB3SNL4P63WUWJ5HY","target":"record","payload":{"canonical_record":{"source":{"id":"2402.09154","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-14T13:13:26Z","cross_cats_sorted":[],"title_canon_sha256":"7fe82e64af51279a4c2fd8c93b64da2f08b17729f9c64bbe6c4ef189e402c6e4","abstract_canon_sha256":"04647cc40c6ec86f14d4147814d65b55465ff8907f31b5965d9b3799842ffaf2"},"schema_version":"1.0"},"canonical_sha256":"f158f188374143b9357c7fb76a593d3e28875063ac105a30263b778314676cfc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:22:44.171312Z","signature_b64":"OisLSYTF2kWaiaN93yJ5mV56AeishRLSbpXySlnqNydffEXTL+g6hMwdWn6N0PWKtn4nUEjzQsStuPvkVIFrDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f158f188374143b9357c7fb76a593d3e28875063ac105a30263b778314676cfc","last_reissued_at":"2026-07-05T10:22:44.170683Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:22:44.170683Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.09154","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:22:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qB2Cwho6JvlNoLzrxR1zI5N+CmpXg2lyWhZorS2UdPQRahidPVm5C1BlJ2Lw0qEv2PL1AJCrSEelyRYJHhClCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T06:22:29.913009Z"},"content_sha256":"4bf1633d4c56ad0765954ea68c86dd59962cec47f2b39a31280d8d16a462d640","schema_version":"1.0","event_id":"sha256:4bf1633d4c56ad0765954ea68c86dd59962cec47f2b39a31280d8d16a462d640"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:6FMPDCBXIFB3SNL4P63WUWJ5HY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Attacking Large Language Models with Projected Gradient Descent","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Johannes Gasteiger, M. H. I. Abdalla, Simon Geisler, Stephan G\\\"unnemann, Tom Wollschl\\\"ager","submitted_at":"2024-02-14T13:13:26Z","abstract_excerpt":"Current LLM alignment methods are readily broken through specifically crafted adversarial prompts. While crafting adversarial prompts using discrete optimization is highly effective, such attacks typically use more than 100,000 LLM calls. This high computational cost makes them unsuitable for, e.g., quantitative analyses and adversarial training. To remedy this, we revisit Projected Gradient Descent (PGD) on the continuously relaxed input prompt. Although previous attempts with ordinary gradient-based attacks largely failed, we show that carefully controlling the error introduced by the contin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.09154","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2402.09154/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:22:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5V8wj4QfcoTKi5YckFOuNYBvANvSt0vpggCMfVJaCZiS9+Exy8CPH6mbs+t1j33gQtMuYlkAg5d8FuFJmAIrAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T06:22:29.913530Z"},"content_sha256":"6fb077fd3d899d5ccc469b5496a1679f169df19b925fd29e1d17a949eb1f1456","schema_version":"1.0","event_id":"sha256:6fb077fd3d899d5ccc469b5496a1679f169df19b925fd29e1d17a949eb1f1456"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6FMPDCBXIFB3SNL4P63WUWJ5HY/bundle.json","state_url":"https://pith.science/pith/6FMPDCBXIFB3SNL4P63WUWJ5HY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6FMPDCBXIFB3SNL4P63WUWJ5HY/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T06:22:29Z","links":{"resolver":"https://pith.science/pith/6FMPDCBXIFB3SNL4P63WUWJ5HY","bundle":"https://pith.science/pith/6FMPDCBXIFB3SNL4P63WUWJ5HY/bundle.json","state":"https://pith.science/pith/6FMPDCBXIFB3SNL4P63WUWJ5HY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6FMPDCBXIFB3SNL4P63WUWJ5HY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:6FMPDCBXIFB3SNL4P63WUWJ5HY","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":"04647cc40c6ec86f14d4147814d65b55465ff8907f31b5965d9b3799842ffaf2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-14T13:13:26Z","title_canon_sha256":"7fe82e64af51279a4c2fd8c93b64da2f08b17729f9c64bbe6c4ef189e402c6e4"},"schema_version":"1.0","source":{"id":"2402.09154","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.09154","created_at":"2026-07-05T10:22:44Z"},{"alias_kind":"arxiv_version","alias_value":"2402.09154v2","created_at":"2026-07-05T10:22:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.09154","created_at":"2026-07-05T10:22:44Z"},{"alias_kind":"pith_short_12","alias_value":"6FMPDCBXIFB3","created_at":"2026-07-05T10:22:44Z"},{"alias_kind":"pith_short_16","alias_value":"6FMPDCBXIFB3SNL4","created_at":"2026-07-05T10:22:44Z"},{"alias_kind":"pith_short_8","alias_value":"6FMPDCBX","created_at":"2026-07-05T10:22:44Z"}],"graph_snapshots":[{"event_id":"sha256:6fb077fd3d899d5ccc469b5496a1679f169df19b925fd29e1d17a949eb1f1456","target":"graph","created_at":"2026-07-05T10:22:44Z","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/2402.09154/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Current LLM alignment methods are readily broken through specifically crafted adversarial prompts. While crafting adversarial prompts using discrete optimization is highly effective, such attacks typically use more than 100,000 LLM calls. This high computational cost makes them unsuitable for, e.g., quantitative analyses and adversarial training. To remedy this, we revisit Projected Gradient Descent (PGD) on the continuously relaxed input prompt. Although previous attempts with ordinary gradient-based attacks largely failed, we show that carefully controlling the error introduced by the contin","authors_text":"Johannes Gasteiger, M. H. I. Abdalla, Simon Geisler, Stephan G\\\"unnemann, Tom Wollschl\\\"ager","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-14T13:13:26Z","title":"Attacking Large Language Models with Projected Gradient Descent"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.09154","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":"sha256:4bf1633d4c56ad0765954ea68c86dd59962cec47f2b39a31280d8d16a462d640","target":"record","created_at":"2026-07-05T10:22:44Z","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":"04647cc40c6ec86f14d4147814d65b55465ff8907f31b5965d9b3799842ffaf2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-14T13:13:26Z","title_canon_sha256":"7fe82e64af51279a4c2fd8c93b64da2f08b17729f9c64bbe6c4ef189e402c6e4"},"schema_version":"1.0","source":{"id":"2402.09154","kind":"arxiv","version":2}},"canonical_sha256":"f158f188374143b9357c7fb76a593d3e28875063ac105a30263b778314676cfc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f158f188374143b9357c7fb76a593d3e28875063ac105a30263b778314676cfc","first_computed_at":"2026-07-05T10:22:44.170683Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:22:44.170683Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"OisLSYTF2kWaiaN93yJ5mV56AeishRLSbpXySlnqNydffEXTL+g6hMwdWn6N0PWKtn4nUEjzQsStuPvkVIFrDw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:22:44.171312Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.09154","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4bf1633d4c56ad0765954ea68c86dd59962cec47f2b39a31280d8d16a462d640","sha256:6fb077fd3d899d5ccc469b5496a1679f169df19b925fd29e1d17a949eb1f1456"],"state_sha256":"c887fe9a79940a9327ccd12fab9c0e5daddb138e6d83a5580f62221f2b4f1441"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AjcJ3F23RwK0LwrJXxU0hlqjUmmvia3OOI7k60wPD0Wc8gSI9EIgmEkFRvUOW+e47wzoIvlDU2fRoWb5sn0BCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T06:22:29.919402Z","bundle_sha256":"6e41174500509f98dcb34099534e4a8ccd1a95ff933452686d0feed286c57563"}}