{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:AFD7XNBY4IQNMLKXWKFD5GMCQL","short_pith_number":"pith:AFD7XNBY","canonical_record":{"source":{"id":"2506.11083","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T09:09:54Z","cross_cats_sorted":[],"title_canon_sha256":"41240b2f143e4dd30719cae5557a1fdb04c4e17a457ef8731c84be5d6a475708","abstract_canon_sha256":"df070a962a116b4b81527cc0beaafb182ec3fccc42bf023f6519a07d50f4305c"},"schema_version":"1.0"},"canonical_sha256":"0147fbb438e220d62d57b28a3e998282cce192853434be67e9cdeb8abfc7f16c","source":{"kind":"arxiv","id":"2506.11083","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.11083","created_at":"2026-06-02T01:03:32Z"},{"alias_kind":"arxiv_version","alias_value":"2506.11083v3","created_at":"2026-06-02T01:03:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11083","created_at":"2026-06-02T01:03:32Z"},{"alias_kind":"pith_short_12","alias_value":"AFD7XNBY4IQN","created_at":"2026-06-02T01:03:32Z"},{"alias_kind":"pith_short_16","alias_value":"AFD7XNBY4IQNMLKX","created_at":"2026-06-02T01:03:32Z"},{"alias_kind":"pith_short_8","alias_value":"AFD7XNBY","created_at":"2026-06-02T01:03:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:AFD7XNBY4IQNMLKXWKFD5GMCQL","target":"record","payload":{"canonical_record":{"source":{"id":"2506.11083","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T09:09:54Z","cross_cats_sorted":[],"title_canon_sha256":"41240b2f143e4dd30719cae5557a1fdb04c4e17a457ef8731c84be5d6a475708","abstract_canon_sha256":"df070a962a116b4b81527cc0beaafb182ec3fccc42bf023f6519a07d50f4305c"},"schema_version":"1.0"},"canonical_sha256":"0147fbb438e220d62d57b28a3e998282cce192853434be67e9cdeb8abfc7f16c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:32.158346Z","signature_b64":"REOs02uVvxf53eovpnJG6eL4mJDh94r2104FL9JRFL/DXFXXRmXqSyWXdoYFG+tGfzcqCzP9ixdtl9XlTJdDBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0147fbb438e220d62d57b28a3e998282cce192853434be67e9cdeb8abfc7f16c","last_reissued_at":"2026-06-02T01:03:32.157663Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:32.157663Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.11083","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:03:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jw2XKuOhEq96fzbIANCyrlLqSdKpzKYFWmblqZ7Z5c9S7B6XDDVy0J+FE0Cgdr6t89xGIbLL9y5awxIIJN8iDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T15:57:29.939641Z"},"content_sha256":"5b951cc96f01e3c283360fccee181d40c7222716474ae0d3d256f1a1cac908aa","schema_version":"1.0","event_id":"sha256:5b951cc96f01e3c283360fccee181d40c7222716474ae0d3d256f1a1cac908aa"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:AFD7XNBY4IQNMLKXWKFD5GMCQL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RedDebate: Safer Responses Through Multi-Agent Red Teaming Debates","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ali Asad, Radin Shayanfar, Stephen Obadinma, Xiaodan Zhu","submitted_at":"2025-06-04T09:09:54Z","abstract_excerpt":"We introduce RedDebate, a novel multi-agent debate framework that provides the foundation for Large Language Models (LLMs) to identify and mitigate their unsafe behaviours. AI safety approaches often rely on costly human evaluation or isolated single-model assessment, both constrained by scalability and prone to oversight failures. RedDebate employs collaborative argumentation among multiple LLMs across diverse debate scenarios, enabling them to critically evaluate one another's reasoning and systematically uncover unsafe failure modes through fully automated red-teaming. To support this, we p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11083","kind":"arxiv","version":3},"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/2506.11083/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":"2026-06-02T01:03:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9PytlxWkSl2ERQ2GshDJ03FhzK1lO64LZf48gml4pyYxMqT3e/icxUlFM36oEq5G/Z3TK+6K8kchx6+hK4GbDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T15:57:29.940021Z"},"content_sha256":"5a81c00ada877f0d5922b69c2b2d2690cbf8b91c679c15ad0749872b171b07ae","schema_version":"1.0","event_id":"sha256:5a81c00ada877f0d5922b69c2b2d2690cbf8b91c679c15ad0749872b171b07ae"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AFD7XNBY4IQNMLKXWKFD5GMCQL/bundle.json","state_url":"https://pith.science/pith/AFD7XNBY4IQNMLKXWKFD5GMCQL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AFD7XNBY4IQNMLKXWKFD5GMCQL/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":"2026-06-05T15:57:29Z","links":{"resolver":"https://pith.science/pith/AFD7XNBY4IQNMLKXWKFD5GMCQL","bundle":"https://pith.science/pith/AFD7XNBY4IQNMLKXWKFD5GMCQL/bundle.json","state":"https://pith.science/pith/AFD7XNBY4IQNMLKXWKFD5GMCQL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AFD7XNBY4IQNMLKXWKFD5GMCQL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:AFD7XNBY4IQNMLKXWKFD5GMCQL","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":"df070a962a116b4b81527cc0beaafb182ec3fccc42bf023f6519a07d50f4305c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T09:09:54Z","title_canon_sha256":"41240b2f143e4dd30719cae5557a1fdb04c4e17a457ef8731c84be5d6a475708"},"schema_version":"1.0","source":{"id":"2506.11083","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.11083","created_at":"2026-06-02T01:03:32Z"},{"alias_kind":"arxiv_version","alias_value":"2506.11083v3","created_at":"2026-06-02T01:03:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11083","created_at":"2026-06-02T01:03:32Z"},{"alias_kind":"pith_short_12","alias_value":"AFD7XNBY4IQN","created_at":"2026-06-02T01:03:32Z"},{"alias_kind":"pith_short_16","alias_value":"AFD7XNBY4IQNMLKX","created_at":"2026-06-02T01:03:32Z"},{"alias_kind":"pith_short_8","alias_value":"AFD7XNBY","created_at":"2026-06-02T01:03:32Z"}],"graph_snapshots":[{"event_id":"sha256:5a81c00ada877f0d5922b69c2b2d2690cbf8b91c679c15ad0749872b171b07ae","target":"graph","created_at":"2026-06-02T01:03:32Z","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/2506.11083/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce RedDebate, a novel multi-agent debate framework that provides the foundation for Large Language Models (LLMs) to identify and mitigate their unsafe behaviours. AI safety approaches often rely on costly human evaluation or isolated single-model assessment, both constrained by scalability and prone to oversight failures. RedDebate employs collaborative argumentation among multiple LLMs across diverse debate scenarios, enabling them to critically evaluate one another's reasoning and systematically uncover unsafe failure modes through fully automated red-teaming. To support this, we p","authors_text":"Ali Asad, Radin Shayanfar, Stephen Obadinma, Xiaodan Zhu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T09:09:54Z","title":"RedDebate: Safer Responses Through Multi-Agent Red Teaming Debates"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11083","kind":"arxiv","version":3},"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":"sha256:5b951cc96f01e3c283360fccee181d40c7222716474ae0d3d256f1a1cac908aa","target":"record","created_at":"2026-06-02T01:03:32Z","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":"df070a962a116b4b81527cc0beaafb182ec3fccc42bf023f6519a07d50f4305c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T09:09:54Z","title_canon_sha256":"41240b2f143e4dd30719cae5557a1fdb04c4e17a457ef8731c84be5d6a475708"},"schema_version":"1.0","source":{"id":"2506.11083","kind":"arxiv","version":3}},"canonical_sha256":"0147fbb438e220d62d57b28a3e998282cce192853434be67e9cdeb8abfc7f16c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0147fbb438e220d62d57b28a3e998282cce192853434be67e9cdeb8abfc7f16c","first_computed_at":"2026-06-02T01:03:32.157663Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:32.157663Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"REOs02uVvxf53eovpnJG6eL4mJDh94r2104FL9JRFL/DXFXXRmXqSyWXdoYFG+tGfzcqCzP9ixdtl9XlTJdDBQ==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:32.158346Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.11083","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5b951cc96f01e3c283360fccee181d40c7222716474ae0d3d256f1a1cac908aa","sha256:5a81c00ada877f0d5922b69c2b2d2690cbf8b91c679c15ad0749872b171b07ae"],"state_sha256":"de0d8cd43662542750f80a61caa2f3e9d190cc4a8cd0a953022769ace3842403"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SllshoW9wleocGnIPHnW6IDxm5mzokPBpDLb01dj9fPoPNH1Hs9YbACOu3DJyeAAPhF5a5Nlh+OKIB9mAcGDCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-05T15:57:29.942069Z","bundle_sha256":"17244670f24b333e46ad8936538480e1aff73330eb41e1572aed747541c4ecfb"}}