{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:UM3JQHL63RQYL5N24SNNFJ7K47","short_pith_number":"pith:UM3JQHL6","canonical_record":{"source":{"id":"2410.21791","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-29T06:54:00Z","cross_cats_sorted":[],"title_canon_sha256":"6cc38b8f1fd44a8b527e6f055bc9db95d35e05fa86d20ede7ea350fb93b13f14","abstract_canon_sha256":"14c360bb076c75e2b3b7e9b8a5fe4ae8dbb4a5eb7729f4abf2023c4ee431307e"},"schema_version":"1.0"},"canonical_sha256":"a336981d7edc6185f5bae49ad2a7eae7f1fb8d2d90240ab0e94fa5611491f5c9","source":{"kind":"arxiv","id":"2410.21791","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.21791","created_at":"2026-07-05T09:27:33Z"},{"alias_kind":"arxiv_version","alias_value":"2410.21791v1","created_at":"2026-07-05T09:27:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.21791","created_at":"2026-07-05T09:27:33Z"},{"alias_kind":"pith_short_12","alias_value":"UM3JQHL63RQY","created_at":"2026-07-05T09:27:33Z"},{"alias_kind":"pith_short_16","alias_value":"UM3JQHL63RQYL5N2","created_at":"2026-07-05T09:27:33Z"},{"alias_kind":"pith_short_8","alias_value":"UM3JQHL6","created_at":"2026-07-05T09:27:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:UM3JQHL63RQYL5N24SNNFJ7K47","target":"record","payload":{"canonical_record":{"source":{"id":"2410.21791","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-29T06:54:00Z","cross_cats_sorted":[],"title_canon_sha256":"6cc38b8f1fd44a8b527e6f055bc9db95d35e05fa86d20ede7ea350fb93b13f14","abstract_canon_sha256":"14c360bb076c75e2b3b7e9b8a5fe4ae8dbb4a5eb7729f4abf2023c4ee431307e"},"schema_version":"1.0"},"canonical_sha256":"a336981d7edc6185f5bae49ad2a7eae7f1fb8d2d90240ab0e94fa5611491f5c9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:33.441029Z","signature_b64":"Hkr+4Dkc6GNKlAfSQwFGvEVzUrf0k35xlZD+JES3Q3DbF7e/2WYG6YFq/2qpw8NF87ROg7IiNHOguxCaqdKRCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a336981d7edc6185f5bae49ad2a7eae7f1fb8d2d90240ab0e94fa5611491f5c9","last_reissued_at":"2026-07-05T09:27:33.440534Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:33.440534Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.21791","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":"2026-07-05T09:27:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IzpMwfNvfwoV5UXtooVG4A+F9FmhoXaUVGLbre63mFcFijNz9bGkUDpR7gfyXB5Vdsduf+chLARr3J9i9HtPDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T10:27:17.901953Z"},"content_sha256":"4e32d6c417c44d706340eb492c13467bab3a704e09c8ac2d558f19f8962d3d96","schema_version":"1.0","event_id":"sha256:4e32d6c417c44d706340eb492c13467bab3a704e09c8ac2d558f19f8962d3d96"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:UM3JQHL63RQYL5N24SNNFJ7K47","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Enhancing Adversarial Attacks through Chain of Thought","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jingbo Su","submitted_at":"2024-10-29T06:54:00Z","abstract_excerpt":"Large language models (LLMs) have demonstrated impressive performance across various domains but remain susceptible to safety concerns. Prior research indicates that gradient-based adversarial attacks are particularly effective against aligned LLMs and the chain of thought (CoT) prompting can elicit desired answers through step-by-step reasoning. This paper proposes enhancing the robustness of adversarial attacks on aligned LLMs by integrating CoT prompts with the greedy coordinate gradient (GCG) technique. Using CoT triggers instead of affirmative targets stimulates the reasoning abilities of"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.21791","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/2410.21791/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":"2026-07-05T09:27:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tuvUa2DJPehVfCZKeYeA2as0jMj/rExHhI1Q7vAC02mxA+FjS78XuzBsVM5sr4KJ/WSyltoEc12XIBywSAV6Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T10:27:17.902888Z"},"content_sha256":"3550000119bb5c49db3578044ea8dd7c80222bdf41c7764db2ba2d65f5b10891","schema_version":"1.0","event_id":"sha256:3550000119bb5c49db3578044ea8dd7c80222bdf41c7764db2ba2d65f5b10891"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UM3JQHL63RQYL5N24SNNFJ7K47/bundle.json","state_url":"https://pith.science/pith/UM3JQHL63RQYL5N24SNNFJ7K47/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UM3JQHL63RQYL5N24SNNFJ7K47/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":"2026-08-15T10:27:17Z","links":{"resolver":"https://pith.science/pith/UM3JQHL63RQYL5N24SNNFJ7K47","bundle":"https://pith.science/pith/UM3JQHL63RQYL5N24SNNFJ7K47/bundle.json","state":"https://pith.science/pith/UM3JQHL63RQYL5N24SNNFJ7K47/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UM3JQHL63RQYL5N24SNNFJ7K47/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:UM3JQHL63RQYL5N24SNNFJ7K47","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":"14c360bb076c75e2b3b7e9b8a5fe4ae8dbb4a5eb7729f4abf2023c4ee431307e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-29T06:54:00Z","title_canon_sha256":"6cc38b8f1fd44a8b527e6f055bc9db95d35e05fa86d20ede7ea350fb93b13f14"},"schema_version":"1.0","source":{"id":"2410.21791","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.21791","created_at":"2026-07-05T09:27:33Z"},{"alias_kind":"arxiv_version","alias_value":"2410.21791v1","created_at":"2026-07-05T09:27:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.21791","created_at":"2026-07-05T09:27:33Z"},{"alias_kind":"pith_short_12","alias_value":"UM3JQHL63RQY","created_at":"2026-07-05T09:27:33Z"},{"alias_kind":"pith_short_16","alias_value":"UM3JQHL63RQYL5N2","created_at":"2026-07-05T09:27:33Z"},{"alias_kind":"pith_short_8","alias_value":"UM3JQHL6","created_at":"2026-07-05T09:27:33Z"}],"graph_snapshots":[{"event_id":"sha256:3550000119bb5c49db3578044ea8dd7c80222bdf41c7764db2ba2d65f5b10891","target":"graph","created_at":"2026-07-05T09:27:33Z","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/2410.21791/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have demonstrated impressive performance across various domains but remain susceptible to safety concerns. Prior research indicates that gradient-based adversarial attacks are particularly effective against aligned LLMs and the chain of thought (CoT) prompting can elicit desired answers through step-by-step reasoning. This paper proposes enhancing the robustness of adversarial attacks on aligned LLMs by integrating CoT prompts with the greedy coordinate gradient (GCG) technique. Using CoT triggers instead of affirmative targets stimulates the reasoning abilities of","authors_text":"Jingbo Su","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-29T06:54:00Z","title":"Enhancing Adversarial Attacks through Chain of Thought"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.21791","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":"sha256:4e32d6c417c44d706340eb492c13467bab3a704e09c8ac2d558f19f8962d3d96","target":"record","created_at":"2026-07-05T09:27:33Z","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":"14c360bb076c75e2b3b7e9b8a5fe4ae8dbb4a5eb7729f4abf2023c4ee431307e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-29T06:54:00Z","title_canon_sha256":"6cc38b8f1fd44a8b527e6f055bc9db95d35e05fa86d20ede7ea350fb93b13f14"},"schema_version":"1.0","source":{"id":"2410.21791","kind":"arxiv","version":1}},"canonical_sha256":"a336981d7edc6185f5bae49ad2a7eae7f1fb8d2d90240ab0e94fa5611491f5c9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a336981d7edc6185f5bae49ad2a7eae7f1fb8d2d90240ab0e94fa5611491f5c9","first_computed_at":"2026-07-05T09:27:33.440534Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:27:33.440534Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Hkr+4Dkc6GNKlAfSQwFGvEVzUrf0k35xlZD+JES3Q3DbF7e/2WYG6YFq/2qpw8NF87ROg7IiNHOguxCaqdKRCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:27:33.441029Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.21791","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4e32d6c417c44d706340eb492c13467bab3a704e09c8ac2d558f19f8962d3d96","sha256:3550000119bb5c49db3578044ea8dd7c80222bdf41c7764db2ba2d65f5b10891"],"state_sha256":"803546054d1f0afe3226bd90c93e5aed004561e964ceeed685de13fb7b5e5b04"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dW4S9ikZzUQs+jrovWyqxNyyHKGym423HSy9EKvpDyha2ggDIZxUFU+h40aGpZ5i213tOtF+ZkbQnxQZQqr7Bw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T10:27:17.908085Z","bundle_sha256":"2a5f07e8b11dc9cee2d09e7fa17ef4af9736310d6d599bfb8c6d5b5c62dac13b"}}