{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:5JZ4ZJ6SUPKCUPHDGU4XTRAUAL","short_pith_number":"pith:5JZ4ZJ6S","canonical_record":{"source":{"id":"2603.24511","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-03-25T16:50:56Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"39c50ccc005fb201f67c2bf0c578f4d0131df32dd6be4640d80a8a2098a35af8","abstract_canon_sha256":"08be702c3fc9e5375a6218721be13105f4cc0a087f4d9a007764f574b08c7c51"},"schema_version":"1.0"},"canonical_sha256":"ea73cca7d2a3d42a3ce3353979c41402fdbdd2b43a779519764b47b3de0d38e7","source":{"kind":"arxiv","id":"2603.24511","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.24511","created_at":"2026-06-02T01:03:15Z"},{"alias_kind":"arxiv_version","alias_value":"2603.24511v2","created_at":"2026-06-02T01:03:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.24511","created_at":"2026-06-02T01:03:15Z"},{"alias_kind":"pith_short_12","alias_value":"5JZ4ZJ6SUPKC","created_at":"2026-06-02T01:03:15Z"},{"alias_kind":"pith_short_16","alias_value":"5JZ4ZJ6SUPKCUPHD","created_at":"2026-06-02T01:03:15Z"},{"alias_kind":"pith_short_8","alias_value":"5JZ4ZJ6S","created_at":"2026-06-02T01:03:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:5JZ4ZJ6SUPKCUPHDGU4XTRAUAL","target":"record","payload":{"canonical_record":{"source":{"id":"2603.24511","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-03-25T16:50:56Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"39c50ccc005fb201f67c2bf0c578f4d0131df32dd6be4640d80a8a2098a35af8","abstract_canon_sha256":"08be702c3fc9e5375a6218721be13105f4cc0a087f4d9a007764f574b08c7c51"},"schema_version":"1.0"},"canonical_sha256":"ea73cca7d2a3d42a3ce3353979c41402fdbdd2b43a779519764b47b3de0d38e7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:15.958689Z","signature_b64":"dkRxrMFTQh1blDcfT0XuqY2kLGjMODAofunkHLaVNTkkQHQsi0A2AeXGbK6fdYKj13pRTtZGWEyXsvWJ/o4iAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ea73cca7d2a3d42a3ce3353979c41402fdbdd2b43a779519764b47b3de0d38e7","last_reissued_at":"2026-06-02T01:03:15.958076Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:15.958076Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2603.24511","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":"2026-06-02T01:03:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SfttVa/V+P3dRJoZvfUfMh3RAqy9tCSNTlxkj09wPrwELwj4fI5SvvO4cvrw/4Z0VOgMqY68uOLP94IRQUtBCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T18:32:59.625630Z"},"content_sha256":"a26a91650f627408929351e750bf235df438add3ccf70ba6414a53d4f44da229","schema_version":"1.0","event_id":"sha256:a26a91650f627408929351e750bf235df438add3ccf70ba6414a53d4f44da229"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:5JZ4ZJ6SUPKCUPHDGU4XTRAUAL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Claudini: Autoresearch Discovers State-of-the-Art Adversarial Attack Algorithms for LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Alexander Panfilov, Igor Shilov, Jonas Geiping, Maksym Andriushchenko, Peter Romov, Yves-Alexandre de Montjoye","submitted_at":"2026-03-25T16:50:56Z","abstract_excerpt":"We show that AI agents are capable of discovering novel algorithms for adversarial attacks against LLMs, advancing the state of the art on white-box jailbreaking and prompt injection evaluations. We deploy frontier agents, such as Claude Code and Codex, in an autoresearch loop with access to a library of 30+ prior methods and an evaluation script with a fixed compute budget. We show this pipeline to be effective in jailbreaking OpenAI's GPT-OSS-Safeguard-20B and in prompt injections against Meta-SecAlign-70B, an adversarially robust model. For GPT-OSS-Safeguard, the best agent-discovered metho"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.24511","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/2603.24511/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":"2026-06-02T01:03:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"X7ZRF8jVyDk8v3hZfSdJV6oebF2eC7kytbJ7ttnodlLZan7dIilfQ7FN74PNul7oKNERmi0aH3fJIcTDLW69CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-06T18:32:59.626225Z"},"content_sha256":"c9b8426fc62ee6039eba20169529aac19d2bf0ab03fb8bf9d25eaa5342edae4b","schema_version":"1.0","event_id":"sha256:c9b8426fc62ee6039eba20169529aac19d2bf0ab03fb8bf9d25eaa5342edae4b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5JZ4ZJ6SUPKCUPHDGU4XTRAUAL/bundle.json","state_url":"https://pith.science/pith/5JZ4ZJ6SUPKCUPHDGU4XTRAUAL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5JZ4ZJ6SUPKCUPHDGU4XTRAUAL/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":"2026-06-06T18:32:59Z","links":{"resolver":"https://pith.science/pith/5JZ4ZJ6SUPKCUPHDGU4XTRAUAL","bundle":"https://pith.science/pith/5JZ4ZJ6SUPKCUPHDGU4XTRAUAL/bundle.json","state":"https://pith.science/pith/5JZ4ZJ6SUPKCUPHDGU4XTRAUAL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5JZ4ZJ6SUPKCUPHDGU4XTRAUAL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:5JZ4ZJ6SUPKCUPHDGU4XTRAUAL","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":"08be702c3fc9e5375a6218721be13105f4cc0a087f4d9a007764f574b08c7c51","cross_cats_sorted":["cs.AI","cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-03-25T16:50:56Z","title_canon_sha256":"39c50ccc005fb201f67c2bf0c578f4d0131df32dd6be4640d80a8a2098a35af8"},"schema_version":"1.0","source":{"id":"2603.24511","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.24511","created_at":"2026-06-02T01:03:15Z"},{"alias_kind":"arxiv_version","alias_value":"2603.24511v2","created_at":"2026-06-02T01:03:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.24511","created_at":"2026-06-02T01:03:15Z"},{"alias_kind":"pith_short_12","alias_value":"5JZ4ZJ6SUPKC","created_at":"2026-06-02T01:03:15Z"},{"alias_kind":"pith_short_16","alias_value":"5JZ4ZJ6SUPKCUPHD","created_at":"2026-06-02T01:03:15Z"},{"alias_kind":"pith_short_8","alias_value":"5JZ4ZJ6S","created_at":"2026-06-02T01:03:15Z"}],"graph_snapshots":[{"event_id":"sha256:c9b8426fc62ee6039eba20169529aac19d2bf0ab03fb8bf9d25eaa5342edae4b","target":"graph","created_at":"2026-06-02T01:03:15Z","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/2603.24511/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We show that AI agents are capable of discovering novel algorithms for adversarial attacks against LLMs, advancing the state of the art on white-box jailbreaking and prompt injection evaluations. We deploy frontier agents, such as Claude Code and Codex, in an autoresearch loop with access to a library of 30+ prior methods and an evaluation script with a fixed compute budget. We show this pipeline to be effective in jailbreaking OpenAI's GPT-OSS-Safeguard-20B and in prompt injections against Meta-SecAlign-70B, an adversarially robust model. For GPT-OSS-Safeguard, the best agent-discovered metho","authors_text":"Alexander Panfilov, Igor Shilov, Jonas Geiping, Maksym Andriushchenko, Peter Romov, Yves-Alexandre de Montjoye","cross_cats":["cs.AI","cs.CR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-03-25T16:50:56Z","title":"Claudini: Autoresearch Discovers State-of-the-Art Adversarial Attack Algorithms for LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.24511","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:a26a91650f627408929351e750bf235df438add3ccf70ba6414a53d4f44da229","target":"record","created_at":"2026-06-02T01:03:15Z","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":"08be702c3fc9e5375a6218721be13105f4cc0a087f4d9a007764f574b08c7c51","cross_cats_sorted":["cs.AI","cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-03-25T16:50:56Z","title_canon_sha256":"39c50ccc005fb201f67c2bf0c578f4d0131df32dd6be4640d80a8a2098a35af8"},"schema_version":"1.0","source":{"id":"2603.24511","kind":"arxiv","version":2}},"canonical_sha256":"ea73cca7d2a3d42a3ce3353979c41402fdbdd2b43a779519764b47b3de0d38e7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ea73cca7d2a3d42a3ce3353979c41402fdbdd2b43a779519764b47b3de0d38e7","first_computed_at":"2026-06-02T01:03:15.958076Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:15.958076Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dkRxrMFTQh1blDcfT0XuqY2kLGjMODAofunkHLaVNTkkQHQsi0A2AeXGbK6fdYKj13pRTtZGWEyXsvWJ/o4iAg==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:15.958689Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.24511","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a26a91650f627408929351e750bf235df438add3ccf70ba6414a53d4f44da229","sha256:c9b8426fc62ee6039eba20169529aac19d2bf0ab03fb8bf9d25eaa5342edae4b"],"state_sha256":"7f052ec65188953d72744046edd9110b07d143b45873609aad3ba531e2ed73ef"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vS3PWfh+NAqmtfvTfIe3qkKxuzStj2BgAg4YMMqDhXW/t1dmrIxYCzuB9txA0EF304c/4j+v55hxbfQ6G2lsBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-06T18:32:59.629239Z","bundle_sha256":"068809b1debb2a58b2e3fb877cd7ed28ac0888786b4f2792692c8bd46fe22a7c"}}