{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:YMGYXB4IAAZ2X2G2I7THUP3TXS","short_pith_number":"pith:YMGYXB4I","canonical_record":{"source":{"id":"2402.18104","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-02-28T06:50:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f1ca726faf240b4b66b3086675dedf2f946a8070f77a599bb5b4bf66b3b792c9","abstract_canon_sha256":"9c004c11704cf386f1a7b6fb670d277181dfffe49430baa587e45e9dd85318ec"},"schema_version":"1.0"},"canonical_sha256":"c30d8b87880033abe8da47e67a3f73bc9cc7a97a8ed73412664873af9d9558f2","source":{"kind":"arxiv","id":"2402.18104","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.18104","created_at":"2026-07-05T08:29:23Z"},{"alias_kind":"arxiv_version","alias_value":"2402.18104v2","created_at":"2026-07-05T08:29:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.18104","created_at":"2026-07-05T08:29:23Z"},{"alias_kind":"pith_short_12","alias_value":"YMGYXB4IAAZ2","created_at":"2026-07-05T08:29:23Z"},{"alias_kind":"pith_short_16","alias_value":"YMGYXB4IAAZ2X2G2","created_at":"2026-07-05T08:29:23Z"},{"alias_kind":"pith_short_8","alias_value":"YMGYXB4I","created_at":"2026-07-05T08:29:23Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:YMGYXB4IAAZ2X2G2I7THUP3TXS","target":"record","payload":{"canonical_record":{"source":{"id":"2402.18104","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-02-28T06:50:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f1ca726faf240b4b66b3086675dedf2f946a8070f77a599bb5b4bf66b3b792c9","abstract_canon_sha256":"9c004c11704cf386f1a7b6fb670d277181dfffe49430baa587e45e9dd85318ec"},"schema_version":"1.0"},"canonical_sha256":"c30d8b87880033abe8da47e67a3f73bc9cc7a97a8ed73412664873af9d9558f2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:29:23.274384Z","signature_b64":"swXoyER6RnC80jAY852vu2+K4mq6rqqpIOKg2UoSOE/84iDVJ1lACzLV1Zwfc+kvXJwzdBfCkcUWJ4jN4QWzDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c30d8b87880033abe8da47e67a3f73bc9cc7a97a8ed73412664873af9d9558f2","last_reissued_at":"2026-07-05T08:29:23.273879Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:29:23.273879Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.18104","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":"2026-07-05T08:29:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7aGA9Ts/6EtLt+x4Un17ANi2JDBBcJvgM7FOW8909q0IOP8rXGkWFsZYhmnsz/jLyfUHMO1GX7FrJa+0yl0nDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T13:03:00.397378Z"},"content_sha256":"2264666a4f48a29f42b83cf5dfe6d92bd1b26063a1834a3f325f3c2757551f7d","schema_version":"1.0","event_id":"sha256:2264666a4f48a29f42b83cf5dfe6d92bd1b26063a1834a3f325f3c2757551f7d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:YMGYXB4IAAZ2X2G2I7THUP3TXS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Making Them Ask and Answer: Jailbreaking Large Language Models in Few Queries via Disguise and Reconstruction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Guozhu Meng, Kai Chen, Tong Liu, Yingjie Zhang, Yinpeng Dong, Zhe Zhao","submitted_at":"2024-02-28T06:50:14Z","abstract_excerpt":"In recent years, large language models (LLMs) have demonstrated notable success across various tasks, but the trustworthiness of LLMs is still an open problem. One specific threat is the potential to generate toxic or harmful responses. Attackers can craft adversarial prompts that induce harmful responses from LLMs. In this work, we pioneer a theoretical foundation in LLMs security by identifying bias vulnerabilities within the safety fine-tuning and design a black-box jailbreak method named DRA (Disguise and Reconstruction Attack), which conceals harmful instructions through disguise and prom"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.18104","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/2402.18104/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":"2026-07-05T08:29:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"C618kscCv9o6KLbBascZJIazdZcW4GA86KgnDyCkdzaQlQqSrAQfHHvcfW45hs3qx+NGD4SQZcNDW2fv8Wu3BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T13:03:00.398093Z"},"content_sha256":"786d9c35830ab08d25b53c547dd054156b30af523ae301eebf3b1bae8acd1707","schema_version":"1.0","event_id":"sha256:786d9c35830ab08d25b53c547dd054156b30af523ae301eebf3b1bae8acd1707"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YMGYXB4IAAZ2X2G2I7THUP3TXS/bundle.json","state_url":"https://pith.science/pith/YMGYXB4IAAZ2X2G2I7THUP3TXS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YMGYXB4IAAZ2X2G2I7THUP3TXS/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":"2026-08-06T13:03:00Z","links":{"resolver":"https://pith.science/pith/YMGYXB4IAAZ2X2G2I7THUP3TXS","bundle":"https://pith.science/pith/YMGYXB4IAAZ2X2G2I7THUP3TXS/bundle.json","state":"https://pith.science/pith/YMGYXB4IAAZ2X2G2I7THUP3TXS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YMGYXB4IAAZ2X2G2I7THUP3TXS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:YMGYXB4IAAZ2X2G2I7THUP3TXS","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":"9c004c11704cf386f1a7b6fb670d277181dfffe49430baa587e45e9dd85318ec","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-02-28T06:50:14Z","title_canon_sha256":"f1ca726faf240b4b66b3086675dedf2f946a8070f77a599bb5b4bf66b3b792c9"},"schema_version":"1.0","source":{"id":"2402.18104","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.18104","created_at":"2026-07-05T08:29:23Z"},{"alias_kind":"arxiv_version","alias_value":"2402.18104v2","created_at":"2026-07-05T08:29:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.18104","created_at":"2026-07-05T08:29:23Z"},{"alias_kind":"pith_short_12","alias_value":"YMGYXB4IAAZ2","created_at":"2026-07-05T08:29:23Z"},{"alias_kind":"pith_short_16","alias_value":"YMGYXB4IAAZ2X2G2","created_at":"2026-07-05T08:29:23Z"},{"alias_kind":"pith_short_8","alias_value":"YMGYXB4I","created_at":"2026-07-05T08:29:23Z"}],"graph_snapshots":[{"event_id":"sha256:786d9c35830ab08d25b53c547dd054156b30af523ae301eebf3b1bae8acd1707","target":"graph","created_at":"2026-07-05T08:29:23Z","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/2402.18104/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In recent years, large language models (LLMs) have demonstrated notable success across various tasks, but the trustworthiness of LLMs is still an open problem. One specific threat is the potential to generate toxic or harmful responses. Attackers can craft adversarial prompts that induce harmful responses from LLMs. In this work, we pioneer a theoretical foundation in LLMs security by identifying bias vulnerabilities within the safety fine-tuning and design a black-box jailbreak method named DRA (Disguise and Reconstruction Attack), which conceals harmful instructions through disguise and prom","authors_text":"Guozhu Meng, Kai Chen, Tong Liu, Yingjie Zhang, Yinpeng Dong, Zhe Zhao","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-02-28T06:50:14Z","title":"Making Them Ask and Answer: Jailbreaking Large Language Models in Few Queries via Disguise and Reconstruction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.18104","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":"sha256:2264666a4f48a29f42b83cf5dfe6d92bd1b26063a1834a3f325f3c2757551f7d","target":"record","created_at":"2026-07-05T08:29:23Z","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":"9c004c11704cf386f1a7b6fb670d277181dfffe49430baa587e45e9dd85318ec","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-02-28T06:50:14Z","title_canon_sha256":"f1ca726faf240b4b66b3086675dedf2f946a8070f77a599bb5b4bf66b3b792c9"},"schema_version":"1.0","source":{"id":"2402.18104","kind":"arxiv","version":2}},"canonical_sha256":"c30d8b87880033abe8da47e67a3f73bc9cc7a97a8ed73412664873af9d9558f2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c30d8b87880033abe8da47e67a3f73bc9cc7a97a8ed73412664873af9d9558f2","first_computed_at":"2026-07-05T08:29:23.273879Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:29:23.273879Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"swXoyER6RnC80jAY852vu2+K4mq6rqqpIOKg2UoSOE/84iDVJ1lACzLV1Zwfc+kvXJwzdBfCkcUWJ4jN4QWzDw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:29:23.274384Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.18104","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2264666a4f48a29f42b83cf5dfe6d92bd1b26063a1834a3f325f3c2757551f7d","sha256:786d9c35830ab08d25b53c547dd054156b30af523ae301eebf3b1bae8acd1707"],"state_sha256":"c47bde8d633dff7c1d650c1a793346c121ff5f28d2ab9794b643750d945e24d4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wlyLSHcSp93EztK+pb+8JFZd8UuNJFFBp3azaxEetZaj4U+vxmv438YWUZK+v8zRjyKhiBZ/KUMaQSydO4m9AQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T13:03:00.404994Z","bundle_sha256":"a8a03910c5cbb31b5aa66f39791614000a1b5fc0c5f2af723ae867fa0e49c781"}}