{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:QQOCOYPKPBPRYBZ4YUDHOKKV7Z","short_pith_number":"pith:QQOCOYPK","canonical_record":{"source":{"id":"2406.13662","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-19T16:09:58Z","cross_cats_sorted":[],"title_canon_sha256":"b608c54c8830b3a49aac1bddcae307c80cd198346a718e8b71a7d80a32b94021","abstract_canon_sha256":"11f3e197f7aaeefa890a95e206071738beb72217b273514fc60cb15549e97d10"},"schema_version":"1.0"},"canonical_sha256":"841c2761ea785f1c073cc506772955fe63744d11d7407547f2787e8312dcbfb3","source":{"kind":"arxiv","id":"2406.13662","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.13662","created_at":"2026-07-05T10:06:04Z"},{"alias_kind":"arxiv_version","alias_value":"2406.13662v2","created_at":"2026-07-05T10:06:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.13662","created_at":"2026-07-05T10:06:04Z"},{"alias_kind":"pith_short_12","alias_value":"QQOCOYPKPBPR","created_at":"2026-07-05T10:06:04Z"},{"alias_kind":"pith_short_16","alias_value":"QQOCOYPKPBPRYBZ4","created_at":"2026-07-05T10:06:04Z"},{"alias_kind":"pith_short_8","alias_value":"QQOCOYPK","created_at":"2026-07-05T10:06:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:QQOCOYPKPBPRYBZ4YUDHOKKV7Z","target":"record","payload":{"canonical_record":{"source":{"id":"2406.13662","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-19T16:09:58Z","cross_cats_sorted":[],"title_canon_sha256":"b608c54c8830b3a49aac1bddcae307c80cd198346a718e8b71a7d80a32b94021","abstract_canon_sha256":"11f3e197f7aaeefa890a95e206071738beb72217b273514fc60cb15549e97d10"},"schema_version":"1.0"},"canonical_sha256":"841c2761ea785f1c073cc506772955fe63744d11d7407547f2787e8312dcbfb3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:06:04.586455Z","signature_b64":"lNrREjz/EH7Of6YIBQjrT9tsGF2fo3VRw8LiMBFuXrPC/YFobgdMtIXojClzw8nmUcVeB5py/VzRL8iWHLN4BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"841c2761ea785f1c073cc506772955fe63744d11d7407547f2787e8312dcbfb3","last_reissued_at":"2026-07-05T10:06:04.585948Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:06:04.585948Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.13662","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":"2026-07-05T10:06:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iR9xUZ/y9DeqceS0lRN1wYpA3Q4VM6XLqGfs1JsPE2wwvvdVDdQlx5RKf926CLNgcYRg42SASV59670bGwiSCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T21:29:00.269662Z"},"content_sha256":"c6ca560f9e6e969ebcd2888e3a13826c75a8d0f7e0b8980fd8e407a042964009","schema_version":"1.0","event_id":"sha256:c6ca560f9e6e969ebcd2888e3a13826c75a8d0f7e0b8980fd8e407a042964009"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:QQOCOYPKPBPRYBZ4YUDHOKKV7Z","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Jailbreaking Large Language Models Through Alignment Vulnerabilities in Out-of-Distribution Settings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bingda Tang, Dongping Chen, Jingyu Tang, Lichao Sun, Philip S. Yu, Xiangliang Zhang, Yao Wan, Yue Huang","submitted_at":"2024-06-19T16:09:58Z","abstract_excerpt":"Recently, Large Language Models (LLMs) have garnered significant attention for their exceptional natural language processing capabilities. However, concerns about their trustworthiness remain unresolved, particularly in addressing ``jailbreaking'' attacks on aligned LLMs. Previous research predominantly relies on scenarios involving white-box LLMs or specific, fixed prompt templates, which are often impractical and lack broad applicability. In this paper, we introduce a straightforward and novel method called ObscurePrompt for jailbreaking LLMs, inspired by the observed fragile alignments in O"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.13662","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/2406.13662/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":"2026-07-05T10:06:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tI7eFG5zWhOoRVlJtKVdhgaGINc40hWj101OrwCguwuluCptcpnRa6W4EeXgTjgSsaFiTGGjpb93pUsrd8ueAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T21:29:00.270207Z"},"content_sha256":"37f55c5e3e777cc96a5ac8dd219216903e2cff419acf0b089fb4d2308c6e3c6f","schema_version":"1.0","event_id":"sha256:37f55c5e3e777cc96a5ac8dd219216903e2cff419acf0b089fb4d2308c6e3c6f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QQOCOYPKPBPRYBZ4YUDHOKKV7Z/bundle.json","state_url":"https://pith.science/pith/QQOCOYPKPBPRYBZ4YUDHOKKV7Z/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QQOCOYPKPBPRYBZ4YUDHOKKV7Z/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":"2026-08-08T21:29:00Z","links":{"resolver":"https://pith.science/pith/QQOCOYPKPBPRYBZ4YUDHOKKV7Z","bundle":"https://pith.science/pith/QQOCOYPKPBPRYBZ4YUDHOKKV7Z/bundle.json","state":"https://pith.science/pith/QQOCOYPKPBPRYBZ4YUDHOKKV7Z/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QQOCOYPKPBPRYBZ4YUDHOKKV7Z/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:QQOCOYPKPBPRYBZ4YUDHOKKV7Z","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":"11f3e197f7aaeefa890a95e206071738beb72217b273514fc60cb15549e97d10","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-19T16:09:58Z","title_canon_sha256":"b608c54c8830b3a49aac1bddcae307c80cd198346a718e8b71a7d80a32b94021"},"schema_version":"1.0","source":{"id":"2406.13662","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.13662","created_at":"2026-07-05T10:06:04Z"},{"alias_kind":"arxiv_version","alias_value":"2406.13662v2","created_at":"2026-07-05T10:06:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.13662","created_at":"2026-07-05T10:06:04Z"},{"alias_kind":"pith_short_12","alias_value":"QQOCOYPKPBPR","created_at":"2026-07-05T10:06:04Z"},{"alias_kind":"pith_short_16","alias_value":"QQOCOYPKPBPRYBZ4","created_at":"2026-07-05T10:06:04Z"},{"alias_kind":"pith_short_8","alias_value":"QQOCOYPK","created_at":"2026-07-05T10:06:04Z"}],"graph_snapshots":[{"event_id":"sha256:37f55c5e3e777cc96a5ac8dd219216903e2cff419acf0b089fb4d2308c6e3c6f","target":"graph","created_at":"2026-07-05T10:06:04Z","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/2406.13662/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recently, Large Language Models (LLMs) have garnered significant attention for their exceptional natural language processing capabilities. However, concerns about their trustworthiness remain unresolved, particularly in addressing ``jailbreaking'' attacks on aligned LLMs. Previous research predominantly relies on scenarios involving white-box LLMs or specific, fixed prompt templates, which are often impractical and lack broad applicability. In this paper, we introduce a straightforward and novel method called ObscurePrompt for jailbreaking LLMs, inspired by the observed fragile alignments in O","authors_text":"Bingda Tang, Dongping Chen, Jingyu Tang, Lichao Sun, Philip S. Yu, Xiangliang Zhang, Yao Wan, Yue Huang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-19T16:09:58Z","title":"Jailbreaking Large Language Models Through Alignment Vulnerabilities in Out-of-Distribution Settings"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.13662","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:c6ca560f9e6e969ebcd2888e3a13826c75a8d0f7e0b8980fd8e407a042964009","target":"record","created_at":"2026-07-05T10:06:04Z","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":"11f3e197f7aaeefa890a95e206071738beb72217b273514fc60cb15549e97d10","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-19T16:09:58Z","title_canon_sha256":"b608c54c8830b3a49aac1bddcae307c80cd198346a718e8b71a7d80a32b94021"},"schema_version":"1.0","source":{"id":"2406.13662","kind":"arxiv","version":2}},"canonical_sha256":"841c2761ea785f1c073cc506772955fe63744d11d7407547f2787e8312dcbfb3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"841c2761ea785f1c073cc506772955fe63744d11d7407547f2787e8312dcbfb3","first_computed_at":"2026-07-05T10:06:04.585948Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:06:04.585948Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lNrREjz/EH7Of6YIBQjrT9tsGF2fo3VRw8LiMBFuXrPC/YFobgdMtIXojClzw8nmUcVeB5py/VzRL8iWHLN4BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:06:04.586455Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.13662","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c6ca560f9e6e969ebcd2888e3a13826c75a8d0f7e0b8980fd8e407a042964009","sha256:37f55c5e3e777cc96a5ac8dd219216903e2cff419acf0b089fb4d2308c6e3c6f"],"state_sha256":"18273a8743e71069039871be02d4e7e12a915ebe5147833f85f65d59abf62e0d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Lqv5tNplrJEohJqYpaPktAQseI1emoNYs4GaxDnb7BlT5xrT4YMfCIlWEXLXO+VUa2k6obD1BgECwc4bVF64Dg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T21:29:00.275300Z","bundle_sha256":"23ca15ba5169794f10fa23cc17b3079a783770af59f96995517e84c68d501d4f"}}