{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ZKBWLQ5PZELTT5NR4SMQNO4HIL","short_pith_number":"pith:ZKBWLQ5P","canonical_record":{"source":{"id":"2503.15754","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-03-20T00:13:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fce990a933647cb8b275076c6710e7f264af00ab58210333b4cebed455d64740","abstract_canon_sha256":"c35942254f88db48b5b6b900c4dddcfc6a6b313c1b2839029478088a08de40dd"},"schema_version":"1.0"},"canonical_sha256":"ca8365c3afc91739f5b1e49906bb8742e894fbb336e72ca6026791db0d4e0857","source":{"kind":"arxiv","id":"2503.15754","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.15754","created_at":"2026-07-05T10:35:58Z"},{"alias_kind":"arxiv_version","alias_value":"2503.15754v1","created_at":"2026-07-05T10:35:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.15754","created_at":"2026-07-05T10:35:58Z"},{"alias_kind":"pith_short_12","alias_value":"ZKBWLQ5PZELT","created_at":"2026-07-05T10:35:58Z"},{"alias_kind":"pith_short_16","alias_value":"ZKBWLQ5PZELTT5NR","created_at":"2026-07-05T10:35:58Z"},{"alias_kind":"pith_short_8","alias_value":"ZKBWLQ5P","created_at":"2026-07-05T10:35:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ZKBWLQ5PZELTT5NR4SMQNO4HIL","target":"record","payload":{"canonical_record":{"source":{"id":"2503.15754","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-03-20T00:13:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fce990a933647cb8b275076c6710e7f264af00ab58210333b4cebed455d64740","abstract_canon_sha256":"c35942254f88db48b5b6b900c4dddcfc6a6b313c1b2839029478088a08de40dd"},"schema_version":"1.0"},"canonical_sha256":"ca8365c3afc91739f5b1e49906bb8742e894fbb336e72ca6026791db0d4e0857","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:35:58.128547Z","signature_b64":"NGQiu43p9rEXXOdknLDE6GmMcBjVulhfhm53lTZX2WTQUOfFak7FM8v9nYeEqtggEHkYZN9j6DMv3CS1IyEhBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ca8365c3afc91739f5b1e49906bb8742e894fbb336e72ca6026791db0d4e0857","last_reissued_at":"2026-07-05T10:35:58.127877Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:35:58.127877Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.15754","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:35:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zjx+cXZxpOx7NtRWo2jD3vfh+jHgQLOn8+CbSXR8ZaaITE5tIInSmOM+sOiIkhmX7W9j5kRnGz0zvMQfNV/5CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T01:27:49.892261Z"},"content_sha256":"adf7b76ccf61ca59cd5281d083726af183134c5f7db5f027f7ebc7ac71d16e7a","schema_version":"1.0","event_id":"sha256:adf7b76ccf61ca59cd5281d083726af183134c5f7db5f027f7ebc7ac71d16e7a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ZKBWLQ5PZELTT5NR4SMQNO4HIL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AutoRedTeamer: Autonomous Red Teaming with Lifelong Attack Integration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Andy Zhou, Bo Li, Francesco Pinto, James Zou, Kevin Wu, Sanmi Koyejo, Shuang Yang, Yi Zeng, Yu Yang, Zhaorun Chen","submitted_at":"2025-03-20T00:13:04Z","abstract_excerpt":"As large language models (LLMs) become increasingly capable, security and safety evaluation are crucial. While current red teaming approaches have made strides in assessing LLM vulnerabilities, they often rely heavily on human input and lack comprehensive coverage of emerging attack vectors. This paper introduces AutoRedTeamer, a novel framework for fully automated, end-to-end red teaming against LLMs. AutoRedTeamer combines a multi-agent architecture with a memory-guided attack selection mechanism to enable continuous discovery and integration of new attack vectors. The dual-agent framework c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.15754","kind":"arxiv","version":1},"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/2503.15754/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":"2026-07-05T10:35:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"02/xzsiUfWtaqyz15TctbrnKoIfRivwd5qk1aHqQYlSVMmTfLAQj9dTYaArzih8/1zuGZHsGGWsHTScxZuf7DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T01:27:49.892648Z"},"content_sha256":"9ea4995c1d60c98681b9a0287ea71d74214f2d13f6d419705c42f0f0e4cc1c52","schema_version":"1.0","event_id":"sha256:9ea4995c1d60c98681b9a0287ea71d74214f2d13f6d419705c42f0f0e4cc1c52"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZKBWLQ5PZELTT5NR4SMQNO4HIL/bundle.json","state_url":"https://pith.science/pith/ZKBWLQ5PZELTT5NR4SMQNO4HIL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZKBWLQ5PZELTT5NR4SMQNO4HIL/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":"2026-08-06T01:27:49Z","links":{"resolver":"https://pith.science/pith/ZKBWLQ5PZELTT5NR4SMQNO4HIL","bundle":"https://pith.science/pith/ZKBWLQ5PZELTT5NR4SMQNO4HIL/bundle.json","state":"https://pith.science/pith/ZKBWLQ5PZELTT5NR4SMQNO4HIL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZKBWLQ5PZELTT5NR4SMQNO4HIL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZKBWLQ5PZELTT5NR4SMQNO4HIL","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":"c35942254f88db48b5b6b900c4dddcfc6a6b313c1b2839029478088a08de40dd","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-03-20T00:13:04Z","title_canon_sha256":"fce990a933647cb8b275076c6710e7f264af00ab58210333b4cebed455d64740"},"schema_version":"1.0","source":{"id":"2503.15754","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.15754","created_at":"2026-07-05T10:35:58Z"},{"alias_kind":"arxiv_version","alias_value":"2503.15754v1","created_at":"2026-07-05T10:35:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.15754","created_at":"2026-07-05T10:35:58Z"},{"alias_kind":"pith_short_12","alias_value":"ZKBWLQ5PZELT","created_at":"2026-07-05T10:35:58Z"},{"alias_kind":"pith_short_16","alias_value":"ZKBWLQ5PZELTT5NR","created_at":"2026-07-05T10:35:58Z"},{"alias_kind":"pith_short_8","alias_value":"ZKBWLQ5P","created_at":"2026-07-05T10:35:58Z"}],"graph_snapshots":[{"event_id":"sha256:9ea4995c1d60c98681b9a0287ea71d74214f2d13f6d419705c42f0f0e4cc1c52","target":"graph","created_at":"2026-07-05T10:35:58Z","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/2503.15754/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) become increasingly capable, security and safety evaluation are crucial. While current red teaming approaches have made strides in assessing LLM vulnerabilities, they often rely heavily on human input and lack comprehensive coverage of emerging attack vectors. This paper introduces AutoRedTeamer, a novel framework for fully automated, end-to-end red teaming against LLMs. AutoRedTeamer combines a multi-agent architecture with a memory-guided attack selection mechanism to enable continuous discovery and integration of new attack vectors. The dual-agent framework c","authors_text":"Andy Zhou, Bo Li, Francesco Pinto, James Zou, Kevin Wu, Sanmi Koyejo, Shuang Yang, Yi Zeng, Yu Yang, Zhaorun Chen","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-03-20T00:13:04Z","title":"AutoRedTeamer: Autonomous Red Teaming with Lifelong Attack Integration"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.15754","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:adf7b76ccf61ca59cd5281d083726af183134c5f7db5f027f7ebc7ac71d16e7a","target":"record","created_at":"2026-07-05T10:35:58Z","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":"c35942254f88db48b5b6b900c4dddcfc6a6b313c1b2839029478088a08de40dd","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-03-20T00:13:04Z","title_canon_sha256":"fce990a933647cb8b275076c6710e7f264af00ab58210333b4cebed455d64740"},"schema_version":"1.0","source":{"id":"2503.15754","kind":"arxiv","version":1}},"canonical_sha256":"ca8365c3afc91739f5b1e49906bb8742e894fbb336e72ca6026791db0d4e0857","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ca8365c3afc91739f5b1e49906bb8742e894fbb336e72ca6026791db0d4e0857","first_computed_at":"2026-07-05T10:35:58.127877Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:35:58.127877Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NGQiu43p9rEXXOdknLDE6GmMcBjVulhfhm53lTZX2WTQUOfFak7FM8v9nYeEqtggEHkYZN9j6DMv3CS1IyEhBg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:35:58.128547Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.15754","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:adf7b76ccf61ca59cd5281d083726af183134c5f7db5f027f7ebc7ac71d16e7a","sha256:9ea4995c1d60c98681b9a0287ea71d74214f2d13f6d419705c42f0f0e4cc1c52"],"state_sha256":"30bb93155581a031419cf004e8bed6a14e9948b087874635fc11cbd1e8a0f454"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Rl51h3S2FdNmgItkg5Aa9Mu4j/Q+O2bQwodZIGB6+isc+WKSO1/tVemhrRGbTyl4p+U9go5RTEhvavNdTD6fDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T01:27:49.895003Z","bundle_sha256":"6644c794d2d332e37d3fdf645523fbb8cdd54d7819064790c6bc40b8fe729322"}}