{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:TH6SPZXZDELM3PQ5YD3RARPAVE","short_pith_number":"pith:TH6SPZXZ","canonical_record":{"source":{"id":"2311.09433","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2023-11-15T23:07:40Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"48869b2ae97ad78a96c2b566ddca4d229cc4c8ea777cd0aa0e44d709a0e80f53","abstract_canon_sha256":"c2897057f81e15eab1aa7568fe0a5f368c7a14847be3c425c9fa93d50b93990a"},"schema_version":"1.0"},"canonical_sha256":"99fd27e6f91916cdbe1dc0f71045e0a9241297fcc7824a2f22896f0a183081df","source":{"kind":"arxiv","id":"2311.09433","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.09433","created_at":"2026-07-05T08:55:49Z"},{"alias_kind":"arxiv_version","alias_value":"2311.09433v3","created_at":"2026-07-05T08:55:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.09433","created_at":"2026-07-05T08:55:49Z"},{"alias_kind":"pith_short_12","alias_value":"TH6SPZXZDELM","created_at":"2026-07-05T08:55:49Z"},{"alias_kind":"pith_short_16","alias_value":"TH6SPZXZDELM3PQ5","created_at":"2026-07-05T08:55:49Z"},{"alias_kind":"pith_short_8","alias_value":"TH6SPZXZ","created_at":"2026-07-05T08:55:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:TH6SPZXZDELM3PQ5YD3RARPAVE","target":"record","payload":{"canonical_record":{"source":{"id":"2311.09433","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2023-11-15T23:07:40Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"48869b2ae97ad78a96c2b566ddca4d229cc4c8ea777cd0aa0e44d709a0e80f53","abstract_canon_sha256":"c2897057f81e15eab1aa7568fe0a5f368c7a14847be3c425c9fa93d50b93990a"},"schema_version":"1.0"},"canonical_sha256":"99fd27e6f91916cdbe1dc0f71045e0a9241297fcc7824a2f22896f0a183081df","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:55:49.185787Z","signature_b64":"pJwCrtLMRwqyeUP0cPWCWS0R7BONO+Vsu3j5Pcus0KsOwi8nDyvoJq7eVMcniZD68aKgHsD0FIT56gTDgTv4Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"99fd27e6f91916cdbe1dc0f71045e0a9241297fcc7824a2f22896f0a183081df","last_reissued_at":"2026-07-05T08:55:49.185350Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:55:49.185350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2311.09433","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":"2026-07-05T08:55:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q1Y2krIuq4laifCCXB/1N8GgS5ZpSNvTaxKr3G9vatZIfn5PNpUJJ+Gnoy+2djnl5aPhlRJMgIHNcPdxaRywDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-16T10:57:19.683623Z"},"content_sha256":"58e11d7edb2033e676335a01f707a181256003123d295b41c3cdb1090e97be23","schema_version":"1.0","event_id":"sha256:58e11d7edb2033e676335a01f707a181256003123d295b41c3cdb1090e97be23"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:TH6SPZXZDELM3PQ5YD3RARPAVE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Trojan Activation Attack: Red-Teaming Large Language Models using Activation Steering for Safety-Alignment","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CR","authors_text":"Haoran Wang, Kai Shu","submitted_at":"2023-11-15T23:07:40Z","abstract_excerpt":"To ensure AI safety, instruction-tuned Large Language Models (LLMs) are specifically trained to ensure alignment, which refers to making models behave in accordance with human intentions. While these models have demonstrated commendable results on various safety benchmarks, the vulnerability of their safety alignment has not been extensively studied. This is particularly troubling given the potential harm that LLMs can inflict. Existing attack methods on LLMs often rely on poisoned training data or the injection of malicious prompts. These approaches compromise the stealthiness and generalizab"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.09433","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/2311.09433/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":"2026-07-05T08:55:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rOzm3nYgI6beu8DIC09RTwx+dqFpEjD1JFKnbiZ16ClUN8esCtOOhVh9/gxFGqQHtAIpuFBSfoZJbnh7Jai5Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-16T10:57:19.684004Z"},"content_sha256":"687d396a1ba5b94b84486f429f1404585ced1a646ce5237fa3a9cb26df59a29b","schema_version":"1.0","event_id":"sha256:687d396a1ba5b94b84486f429f1404585ced1a646ce5237fa3a9cb26df59a29b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TH6SPZXZDELM3PQ5YD3RARPAVE/bundle.json","state_url":"https://pith.science/pith/TH6SPZXZDELM3PQ5YD3RARPAVE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TH6SPZXZDELM3PQ5YD3RARPAVE/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":"2026-07-16T10:57:19Z","links":{"resolver":"https://pith.science/pith/TH6SPZXZDELM3PQ5YD3RARPAVE","bundle":"https://pith.science/pith/TH6SPZXZDELM3PQ5YD3RARPAVE/bundle.json","state":"https://pith.science/pith/TH6SPZXZDELM3PQ5YD3RARPAVE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TH6SPZXZDELM3PQ5YD3RARPAVE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:TH6SPZXZDELM3PQ5YD3RARPAVE","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":"c2897057f81e15eab1aa7568fe0a5f368c7a14847be3c425c9fa93d50b93990a","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2023-11-15T23:07:40Z","title_canon_sha256":"48869b2ae97ad78a96c2b566ddca4d229cc4c8ea777cd0aa0e44d709a0e80f53"},"schema_version":"1.0","source":{"id":"2311.09433","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.09433","created_at":"2026-07-05T08:55:49Z"},{"alias_kind":"arxiv_version","alias_value":"2311.09433v3","created_at":"2026-07-05T08:55:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.09433","created_at":"2026-07-05T08:55:49Z"},{"alias_kind":"pith_short_12","alias_value":"TH6SPZXZDELM","created_at":"2026-07-05T08:55:49Z"},{"alias_kind":"pith_short_16","alias_value":"TH6SPZXZDELM3PQ5","created_at":"2026-07-05T08:55:49Z"},{"alias_kind":"pith_short_8","alias_value":"TH6SPZXZ","created_at":"2026-07-05T08:55:49Z"}],"graph_snapshots":[{"event_id":"sha256:687d396a1ba5b94b84486f429f1404585ced1a646ce5237fa3a9cb26df59a29b","target":"graph","created_at":"2026-07-05T08:55:49Z","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/2311.09433/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"To ensure AI safety, instruction-tuned Large Language Models (LLMs) are specifically trained to ensure alignment, which refers to making models behave in accordance with human intentions. While these models have demonstrated commendable results on various safety benchmarks, the vulnerability of their safety alignment has not been extensively studied. This is particularly troubling given the potential harm that LLMs can inflict. Existing attack methods on LLMs often rely on poisoned training data or the injection of malicious prompts. These approaches compromise the stealthiness and generalizab","authors_text":"Haoran Wang, Kai Shu","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2023-11-15T23:07:40Z","title":"Trojan Activation Attack: Red-Teaming Large Language Models using Activation Steering for Safety-Alignment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.09433","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":"sha256:58e11d7edb2033e676335a01f707a181256003123d295b41c3cdb1090e97be23","target":"record","created_at":"2026-07-05T08:55:49Z","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":"c2897057f81e15eab1aa7568fe0a5f368c7a14847be3c425c9fa93d50b93990a","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2023-11-15T23:07:40Z","title_canon_sha256":"48869b2ae97ad78a96c2b566ddca4d229cc4c8ea777cd0aa0e44d709a0e80f53"},"schema_version":"1.0","source":{"id":"2311.09433","kind":"arxiv","version":3}},"canonical_sha256":"99fd27e6f91916cdbe1dc0f71045e0a9241297fcc7824a2f22896f0a183081df","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"99fd27e6f91916cdbe1dc0f71045e0a9241297fcc7824a2f22896f0a183081df","first_computed_at":"2026-07-05T08:55:49.185350Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:55:49.185350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pJwCrtLMRwqyeUP0cPWCWS0R7BONO+Vsu3j5Pcus0KsOwi8nDyvoJq7eVMcniZD68aKgHsD0FIT56gTDgTv4Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:55:49.185787Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.09433","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:58e11d7edb2033e676335a01f707a181256003123d295b41c3cdb1090e97be23","sha256:687d396a1ba5b94b84486f429f1404585ced1a646ce5237fa3a9cb26df59a29b"],"state_sha256":"4dc90f910a6a4a74aa237c9bb62d663cc5443e5e40f915267b8b86dfafe3a8b0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZVc4Ywe8P/cyveFcmKOPqt0FwejEZAfUPAIkixt8Bcdn0TUTUqSu7xpU130oNtqNT1k2/gVdkVDy+MJSOrPiBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-16T10:57:19.686239Z","bundle_sha256":"9bd805fb87f1c37d51a87f159b13eaac4b4c6adc1e36cbf5de03cb693ac5cd29"}}