{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:QWAGRCG4YQ62TQBUMWZ5ZF7QJW","short_pith_number":"pith:QWAGRCG4","canonical_record":{"source":{"id":"2501.18626","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-01-27T12:48:47Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"15196ebe5e9fd0e9c54769a5ae16c9d302c5133b3fdd11676d13952fc6df85f0","abstract_canon_sha256":"a426b3112b65952092d95226c889289af8cab6066b3dc31c671ac2e9afeea74a"},"schema_version":"1.0"},"canonical_sha256":"85806888dcc43da9c03465b3dc97f04d91bcb162ea7a5d8ee2654595722aa296","source":{"kind":"arxiv","id":"2501.18626","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.18626","created_at":"2026-07-05T11:13:19Z"},{"alias_kind":"arxiv_version","alias_value":"2501.18626v4","created_at":"2026-07-05T11:13:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.18626","created_at":"2026-07-05T11:13:19Z"},{"alias_kind":"pith_short_12","alias_value":"QWAGRCG4YQ62","created_at":"2026-07-05T11:13:19Z"},{"alias_kind":"pith_short_16","alias_value":"QWAGRCG4YQ62TQBU","created_at":"2026-07-05T11:13:19Z"},{"alias_kind":"pith_short_8","alias_value":"QWAGRCG4","created_at":"2026-07-05T11:13:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:QWAGRCG4YQ62TQBUMWZ5ZF7QJW","target":"record","payload":{"canonical_record":{"source":{"id":"2501.18626","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-01-27T12:48:47Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"15196ebe5e9fd0e9c54769a5ae16c9d302c5133b3fdd11676d13952fc6df85f0","abstract_canon_sha256":"a426b3112b65952092d95226c889289af8cab6066b3dc31c671ac2e9afeea74a"},"schema_version":"1.0"},"canonical_sha256":"85806888dcc43da9c03465b3dc97f04d91bcb162ea7a5d8ee2654595722aa296","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:19.502911Z","signature_b64":"KPKX4iMClWB6Yx/b9uBKvlfk0m/Y46B8fIJKoIWJxEuW88NtMNk9YsWGqndoVcZUahOmIMTn1qzuyHt6WyiUCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"85806888dcc43da9c03465b3dc97f04d91bcb162ea7a5d8ee2654595722aa296","last_reissued_at":"2026-07-05T11:13:19.502488Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:19.502488Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.18626","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:13:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iLyOGJnIUEp+7RlX77RixEJeuMs0zm8ORPWFVOERNXRkGwSZ1hPjl7Erp/ZvuwAKXsMi8/fJg6c6xKCvHMiFDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T04:16:23.187616Z"},"content_sha256":"e7b85a208db17e5a66427970f585a40c6d4e3505d3ca252ef7a6b56f3a099ed1","schema_version":"1.0","event_id":"sha256:e7b85a208db17e5a66427970f585a40c6d4e3505d3ca252ef7a6b56f3a099ed1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:QWAGRCG4YQ62TQBUMWZ5ZF7QJW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CR","authors_text":"Noel Crespi, Reza Farahbakhsh, Sergey Berezin","submitted_at":"2025-01-27T12:48:47Z","abstract_excerpt":"We present a novel class of jailbreak adversarial attacks on LLMs, termed Task-in-Prompt (TIP) attacks. Our approach embeds sequence-to-sequence tasks (e.g., cipher decoding, riddles, code execution) into the model's prompt to indirectly generate prohibited inputs. To systematically assess the effectiveness of these attacks, we introduce the PHRYGE benchmark. We demonstrate that our techniques successfully circumvent safeguards in six state-of-the-art language models, including GPT-4o and LLaMA 3.2. Our findings highlight critical weaknesses in current LLM safety alignments and underscore the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.18626","kind":"arxiv","version":4},"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/2501.18626/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":"2026-07-05T11:13:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gTgjfam1RdBMphyFH5FWGwWw7W4wgsaZc+7JupK6lWfWKHRPWD8GG6FczKCBaMqcDNfMdlzDQCDJuBXaQyDBDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T04:16:23.188121Z"},"content_sha256":"9645df80ac6160f2217686158bf2fc0c71eb9c9b706ae3ba691c3135c402ef63","schema_version":"1.0","event_id":"sha256:9645df80ac6160f2217686158bf2fc0c71eb9c9b706ae3ba691c3135c402ef63"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QWAGRCG4YQ62TQBUMWZ5ZF7QJW/bundle.json","state_url":"https://pith.science/pith/QWAGRCG4YQ62TQBUMWZ5ZF7QJW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QWAGRCG4YQ62TQBUMWZ5ZF7QJW/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":"2026-08-06T04:16:23Z","links":{"resolver":"https://pith.science/pith/QWAGRCG4YQ62TQBUMWZ5ZF7QJW","bundle":"https://pith.science/pith/QWAGRCG4YQ62TQBUMWZ5ZF7QJW/bundle.json","state":"https://pith.science/pith/QWAGRCG4YQ62TQBUMWZ5ZF7QJW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QWAGRCG4YQ62TQBUMWZ5ZF7QJW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:QWAGRCG4YQ62TQBUMWZ5ZF7QJW","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":"a426b3112b65952092d95226c889289af8cab6066b3dc31c671ac2e9afeea74a","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-01-27T12:48:47Z","title_canon_sha256":"15196ebe5e9fd0e9c54769a5ae16c9d302c5133b3fdd11676d13952fc6df85f0"},"schema_version":"1.0","source":{"id":"2501.18626","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.18626","created_at":"2026-07-05T11:13:19Z"},{"alias_kind":"arxiv_version","alias_value":"2501.18626v4","created_at":"2026-07-05T11:13:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.18626","created_at":"2026-07-05T11:13:19Z"},{"alias_kind":"pith_short_12","alias_value":"QWAGRCG4YQ62","created_at":"2026-07-05T11:13:19Z"},{"alias_kind":"pith_short_16","alias_value":"QWAGRCG4YQ62TQBU","created_at":"2026-07-05T11:13:19Z"},{"alias_kind":"pith_short_8","alias_value":"QWAGRCG4","created_at":"2026-07-05T11:13:19Z"}],"graph_snapshots":[{"event_id":"sha256:9645df80ac6160f2217686158bf2fc0c71eb9c9b706ae3ba691c3135c402ef63","target":"graph","created_at":"2026-07-05T11:13:19Z","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/2501.18626/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present a novel class of jailbreak adversarial attacks on LLMs, termed Task-in-Prompt (TIP) attacks. Our approach embeds sequence-to-sequence tasks (e.g., cipher decoding, riddles, code execution) into the model's prompt to indirectly generate prohibited inputs. To systematically assess the effectiveness of these attacks, we introduce the PHRYGE benchmark. We demonstrate that our techniques successfully circumvent safeguards in six state-of-the-art language models, including GPT-4o and LLaMA 3.2. Our findings highlight critical weaknesses in current LLM safety alignments and underscore the ","authors_text":"Noel Crespi, Reza Farahbakhsh, Sergey Berezin","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-01-27T12:48:47Z","title":"The TIP of the Iceberg: Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.18626","kind":"arxiv","version":4},"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:e7b85a208db17e5a66427970f585a40c6d4e3505d3ca252ef7a6b56f3a099ed1","target":"record","created_at":"2026-07-05T11:13:19Z","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":"a426b3112b65952092d95226c889289af8cab6066b3dc31c671ac2e9afeea74a","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-01-27T12:48:47Z","title_canon_sha256":"15196ebe5e9fd0e9c54769a5ae16c9d302c5133b3fdd11676d13952fc6df85f0"},"schema_version":"1.0","source":{"id":"2501.18626","kind":"arxiv","version":4}},"canonical_sha256":"85806888dcc43da9c03465b3dc97f04d91bcb162ea7a5d8ee2654595722aa296","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"85806888dcc43da9c03465b3dc97f04d91bcb162ea7a5d8ee2654595722aa296","first_computed_at":"2026-07-05T11:13:19.502488Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:19.502488Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KPKX4iMClWB6Yx/b9uBKvlfk0m/Y46B8fIJKoIWJxEuW88NtMNk9YsWGqndoVcZUahOmIMTn1qzuyHt6WyiUCg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:19.502911Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.18626","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e7b85a208db17e5a66427970f585a40c6d4e3505d3ca252ef7a6b56f3a099ed1","sha256:9645df80ac6160f2217686158bf2fc0c71eb9c9b706ae3ba691c3135c402ef63"],"state_sha256":"a3d7c1aef1173b204c97587ed75f70528e65ecf1058423d9dc2be3160e893c6f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4D+p2hNzcHQRgQOwyIoXpzqLYBmVqGs8SLGfulv3S6ChZB7J1Cu3eZd0Bg5B00czHMyxlRS/zAAmpU6eaifSAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T04:16:23.192111Z","bundle_sha256":"2279575377fa3e3b6da8ab8b3c688cdd01938d11b20272130979d0811ad28076"}}