{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GLGLNX55XBTYX4UXZ5II3CDH7Y","short_pith_number":"pith:GLGLNX55","canonical_record":{"source":{"id":"2408.02632","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-08-05T16:55:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"764509b81657611b4cdee1fb775dc02e556c45b5b7cefe3f035efedca3c2b47c","abstract_canon_sha256":"b636d2fd9611bc06a3fe18afa6a5b0125f0ee78f911c989ba7aa93753cdfa91b"},"schema_version":"1.0"},"canonical_sha256":"32ccb6dfbdb8678bf297cf508d8867fe127420ef100195342e73012f87ac9aed","source":{"kind":"arxiv","id":"2408.02632","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.02632","created_at":"2026-07-05T09:53:03Z"},{"alias_kind":"arxiv_version","alias_value":"2408.02632v2","created_at":"2026-07-05T09:53:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.02632","created_at":"2026-07-05T09:53:03Z"},{"alias_kind":"pith_short_12","alias_value":"GLGLNX55XBTY","created_at":"2026-07-05T09:53:03Z"},{"alias_kind":"pith_short_16","alias_value":"GLGLNX55XBTYX4UX","created_at":"2026-07-05T09:53:03Z"},{"alias_kind":"pith_short_8","alias_value":"GLGLNX55","created_at":"2026-07-05T09:53:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GLGLNX55XBTYX4UXZ5II3CDH7Y","target":"record","payload":{"canonical_record":{"source":{"id":"2408.02632","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-08-05T16:55:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"764509b81657611b4cdee1fb775dc02e556c45b5b7cefe3f035efedca3c2b47c","abstract_canon_sha256":"b636d2fd9611bc06a3fe18afa6a5b0125f0ee78f911c989ba7aa93753cdfa91b"},"schema_version":"1.0"},"canonical_sha256":"32ccb6dfbdb8678bf297cf508d8867fe127420ef100195342e73012f87ac9aed","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:53:03.409855Z","signature_b64":"lBL98BKEi4TCSV54mRnK99yWnOHLMl+TPdIMZjQTf9GvwgMNbzHuzo9aFngMK9deMiB8KoQ9O+TdnDOMBpCVAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"32ccb6dfbdb8678bf297cf508d8867fe127420ef100195342e73012f87ac9aed","last_reissued_at":"2026-07-05T09:53:03.409390Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:53:03.409390Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2408.02632","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":"2026-07-05T09:53:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oXYcUa/Usp3WkU5peerKOFpa2r8pKHMygwQnNDQn53mSWpenQU+V14YPjWyuISLmr1FzemF3c5+17rltYijQDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T17:45:36.686243Z"},"content_sha256":"269abd278cb405ca445a16ed6e691bccb86c67b726b66796ebbb5d988ad24486","schema_version":"1.0","event_id":"sha256:269abd278cb405ca445a16ed6e691bccb86c67b726b66796ebbb5d988ad24486"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GLGLNX55XBTYX4UXZ5II3CDH7Y","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SEAS: Self-Evolving Adversarial Safety Optimization for Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Guogang Liao, Jingang Wang, Muxi Diao, Rumei Li, Shiyang Liu, Weiran Xu, Xunliang Cai","submitted_at":"2024-08-05T16:55:06Z","abstract_excerpt":"As large language models (LLMs) continue to advance in capability and influence, ensuring their security and preventing harmful outputs has become crucial. A promising approach to address these concerns involves training models to automatically generate adversarial prompts for red teaming. However, the evolving subtlety of vulnerabilities in LLMs challenges the effectiveness of current adversarial methods, which struggle to specifically target and explore the weaknesses of these models. To tackle these challenges, we introduce the $\\mathbf{S}\\text{elf-}\\mathbf{E}\\text{volving }\\mathbf{A}\\text{"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.02632","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/2408.02632/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":"2026-07-05T09:53:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fLX7cVJCEXLeOc4S/wTiysGmuG+GmYULIL6wtU69BcQE3jYT1oq+oPdX5j4SNweA7FO5Lfiu60vOvpsjqkAYBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T17:45:36.686623Z"},"content_sha256":"fd97221418e819103ac43bdba6067b01300bc9295791269c01b5568eba9cb08c","schema_version":"1.0","event_id":"sha256:fd97221418e819103ac43bdba6067b01300bc9295791269c01b5568eba9cb08c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GLGLNX55XBTYX4UXZ5II3CDH7Y/bundle.json","state_url":"https://pith.science/pith/GLGLNX55XBTYX4UXZ5II3CDH7Y/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GLGLNX55XBTYX4UXZ5II3CDH7Y/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":"2026-07-06T17:45:36Z","links":{"resolver":"https://pith.science/pith/GLGLNX55XBTYX4UXZ5II3CDH7Y","bundle":"https://pith.science/pith/GLGLNX55XBTYX4UXZ5II3CDH7Y/bundle.json","state":"https://pith.science/pith/GLGLNX55XBTYX4UXZ5II3CDH7Y/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GLGLNX55XBTYX4UXZ5II3CDH7Y/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GLGLNX55XBTYX4UXZ5II3CDH7Y","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":"b636d2fd9611bc06a3fe18afa6a5b0125f0ee78f911c989ba7aa93753cdfa91b","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-08-05T16:55:06Z","title_canon_sha256":"764509b81657611b4cdee1fb775dc02e556c45b5b7cefe3f035efedca3c2b47c"},"schema_version":"1.0","source":{"id":"2408.02632","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.02632","created_at":"2026-07-05T09:53:03Z"},{"alias_kind":"arxiv_version","alias_value":"2408.02632v2","created_at":"2026-07-05T09:53:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.02632","created_at":"2026-07-05T09:53:03Z"},{"alias_kind":"pith_short_12","alias_value":"GLGLNX55XBTY","created_at":"2026-07-05T09:53:03Z"},{"alias_kind":"pith_short_16","alias_value":"GLGLNX55XBTYX4UX","created_at":"2026-07-05T09:53:03Z"},{"alias_kind":"pith_short_8","alias_value":"GLGLNX55","created_at":"2026-07-05T09:53:03Z"}],"graph_snapshots":[{"event_id":"sha256:fd97221418e819103ac43bdba6067b01300bc9295791269c01b5568eba9cb08c","target":"graph","created_at":"2026-07-05T09:53:03Z","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/2408.02632/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) continue to advance in capability and influence, ensuring their security and preventing harmful outputs has become crucial. A promising approach to address these concerns involves training models to automatically generate adversarial prompts for red teaming. However, the evolving subtlety of vulnerabilities in LLMs challenges the effectiveness of current adversarial methods, which struggle to specifically target and explore the weaknesses of these models. To tackle these challenges, we introduce the $\\mathbf{S}\\text{elf-}\\mathbf{E}\\text{volving }\\mathbf{A}\\text{","authors_text":"Guogang Liao, Jingang Wang, Muxi Diao, Rumei Li, Shiyang Liu, Weiran Xu, Xunliang Cai","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-08-05T16:55:06Z","title":"SEAS: Self-Evolving Adversarial Safety Optimization for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.02632","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":"sha256:269abd278cb405ca445a16ed6e691bccb86c67b726b66796ebbb5d988ad24486","target":"record","created_at":"2026-07-05T09:53:03Z","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":"b636d2fd9611bc06a3fe18afa6a5b0125f0ee78f911c989ba7aa93753cdfa91b","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-08-05T16:55:06Z","title_canon_sha256":"764509b81657611b4cdee1fb775dc02e556c45b5b7cefe3f035efedca3c2b47c"},"schema_version":"1.0","source":{"id":"2408.02632","kind":"arxiv","version":2}},"canonical_sha256":"32ccb6dfbdb8678bf297cf508d8867fe127420ef100195342e73012f87ac9aed","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"32ccb6dfbdb8678bf297cf508d8867fe127420ef100195342e73012f87ac9aed","first_computed_at":"2026-07-05T09:53:03.409390Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:53:03.409390Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lBL98BKEi4TCSV54mRnK99yWnOHLMl+TPdIMZjQTf9GvwgMNbzHuzo9aFngMK9deMiB8KoQ9O+TdnDOMBpCVAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:53:03.409855Z","signed_message":"canonical_sha256_bytes"},"source_id":"2408.02632","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:269abd278cb405ca445a16ed6e691bccb86c67b726b66796ebbb5d988ad24486","sha256:fd97221418e819103ac43bdba6067b01300bc9295791269c01b5568eba9cb08c"],"state_sha256":"41d0a7bf5f1316406003d7c41e2e6ded5777a45ceb03a0d4fc105fe7bd65aff2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kK25PFDO212Nksmmxr80uMnejCD2I33ac7quNzA3pf+XO4AJZuLsy48C72vR503aKY+NtRMeBFQx5i+rBNdCDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T17:45:36.689227Z","bundle_sha256":"5f9c6e62090a1235fb623596e65741dbb31c0928167c8a427ddb47c02160f03c"}}