{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:OI2AR2HMJMISZCPRDN7VE6YVVJ","short_pith_number":"pith:OI2AR2HM","canonical_record":{"source":{"id":"2602.09629","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2026-02-10T10:17:25Z","cross_cats_sorted":["cs.AI","cs.CY","cs.ET","cs.HC"],"title_canon_sha256":"4c9a04e9b2f581009370ff62e16fc0802f3d6ae5f092455b665ec5cd29c858e4","abstract_canon_sha256":"94132c4198a4ee3f51ea0196430a577b53033a427e2e89a1260a85828bf9b8a1"},"schema_version":"1.0"},"canonical_sha256":"723408e8ec4b112c89f11b7f527b15aa61902ad49758b72c43c15387afaba5d3","source":{"kind":"arxiv","id":"2602.09629","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.09629","created_at":"2026-06-23T00:11:47Z"},{"alias_kind":"arxiv_version","alias_value":"2602.09629v1","created_at":"2026-06-23T00:11:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.09629","created_at":"2026-06-23T00:11:47Z"},{"alias_kind":"pith_short_12","alias_value":"OI2AR2HMJMIS","created_at":"2026-06-23T00:11:47Z"},{"alias_kind":"pith_short_16","alias_value":"OI2AR2HMJMISZCPR","created_at":"2026-06-23T00:11:47Z"},{"alias_kind":"pith_short_8","alias_value":"OI2AR2HM","created_at":"2026-06-23T00:11:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:OI2AR2HMJMISZCPRDN7VE6YVVJ","target":"record","payload":{"canonical_record":{"source":{"id":"2602.09629","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2026-02-10T10:17:25Z","cross_cats_sorted":["cs.AI","cs.CY","cs.ET","cs.HC"],"title_canon_sha256":"4c9a04e9b2f581009370ff62e16fc0802f3d6ae5f092455b665ec5cd29c858e4","abstract_canon_sha256":"94132c4198a4ee3f51ea0196430a577b53033a427e2e89a1260a85828bf9b8a1"},"schema_version":"1.0"},"canonical_sha256":"723408e8ec4b112c89f11b7f527b15aa61902ad49758b72c43c15387afaba5d3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T00:11:47.079593Z","signature_b64":"rTxL2R9zM+7mPsd2A+2dIj45DdjI1aN52k5YhnZMeHYfAa3T/neR+FOAUaFcpmouS92SmCCSC3mpmKcucMnFCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"723408e8ec4b112c89f11b7f527b15aa61902ad49758b72c43c15387afaba5d3","last_reissued_at":"2026-06-23T00:11:47.079050Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T00:11:47.079050Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2602.09629","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":"2026-06-23T00:11:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EjHPCW0WUMzOeIjL8b1OM44Z7uhbztNWXfC/LONv4ljlz9WFq2mqPvIr34akFEFnQjtdQo6Wh6pVc3KDAhxjCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-24T18:55:17.295546Z"},"content_sha256":"ca18d50df0a61c55217b241d8fbee816f5c9b4f89104f0c1b267abe1e65ca6c8","schema_version":"1.0","event_id":"sha256:ca18d50df0a61c55217b241d8fbee816f5c9b4f89104f0c1b267abe1e65ca6c8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:OI2AR2HMJMISZCPRDN7VE6YVVJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Stop Testing Attacks, Start Diagnosing Defenses: The Four-Checkpoint Framework Reveals Where LLM Safety Breaks","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CY","cs.ET","cs.HC"],"primary_cat":"cs.CR","authors_text":"Hayfa Dhahbi, Kashyap Thimmaraju","submitted_at":"2026-02-10T10:17:25Z","abstract_excerpt":"Large Language Models (LLMs) deploy safety mechanisms to prevent harmful outputs, yet these defenses remain vulnerable to adversarial prompts. While existing research demonstrates that jailbreak attacks succeed, it does not explain \\textit{where} defenses fail or \\textit{why}.\n  To address this gap, we propose that LLM safety operates as a sequential pipeline with distinct checkpoints. We introduce the \\textbf{Four-Checkpoint Framework}, which organizes safety mechanisms along two dimensions: processing stage (input vs.\\ output) and detection level (literal vs.\\ intent). This creates four chec"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.09629","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/2602.09629/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":"2026-06-23T00:11:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"a6SUaGfiiNIRQJNqqtWigMT0s44ZF9UtbYvoJ3qmKVrmVcX04YLvyLvQNCxygUroCPUdLozDFumlV9uWJHuRCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-24T18:55:17.295920Z"},"content_sha256":"dee1bf7bde4a2b4870926d53684e1c9ca774b7be4d25b9c5032b28dbfef9061c","schema_version":"1.0","event_id":"sha256:dee1bf7bde4a2b4870926d53684e1c9ca774b7be4d25b9c5032b28dbfef9061c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OI2AR2HMJMISZCPRDN7VE6YVVJ/bundle.json","state_url":"https://pith.science/pith/OI2AR2HMJMISZCPRDN7VE6YVVJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OI2AR2HMJMISZCPRDN7VE6YVVJ/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":"2026-06-24T18:55:17Z","links":{"resolver":"https://pith.science/pith/OI2AR2HMJMISZCPRDN7VE6YVVJ","bundle":"https://pith.science/pith/OI2AR2HMJMISZCPRDN7VE6YVVJ/bundle.json","state":"https://pith.science/pith/OI2AR2HMJMISZCPRDN7VE6YVVJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OI2AR2HMJMISZCPRDN7VE6YVVJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:OI2AR2HMJMISZCPRDN7VE6YVVJ","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":"94132c4198a4ee3f51ea0196430a577b53033a427e2e89a1260a85828bf9b8a1","cross_cats_sorted":["cs.AI","cs.CY","cs.ET","cs.HC"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2026-02-10T10:17:25Z","title_canon_sha256":"4c9a04e9b2f581009370ff62e16fc0802f3d6ae5f092455b665ec5cd29c858e4"},"schema_version":"1.0","source":{"id":"2602.09629","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.09629","created_at":"2026-06-23T00:11:47Z"},{"alias_kind":"arxiv_version","alias_value":"2602.09629v1","created_at":"2026-06-23T00:11:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.09629","created_at":"2026-06-23T00:11:47Z"},{"alias_kind":"pith_short_12","alias_value":"OI2AR2HMJMIS","created_at":"2026-06-23T00:11:47Z"},{"alias_kind":"pith_short_16","alias_value":"OI2AR2HMJMISZCPR","created_at":"2026-06-23T00:11:47Z"},{"alias_kind":"pith_short_8","alias_value":"OI2AR2HM","created_at":"2026-06-23T00:11:47Z"}],"graph_snapshots":[{"event_id":"sha256:dee1bf7bde4a2b4870926d53684e1c9ca774b7be4d25b9c5032b28dbfef9061c","target":"graph","created_at":"2026-06-23T00:11:47Z","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/2602.09629/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) deploy safety mechanisms to prevent harmful outputs, yet these defenses remain vulnerable to adversarial prompts. While existing research demonstrates that jailbreak attacks succeed, it does not explain \\textit{where} defenses fail or \\textit{why}.\n  To address this gap, we propose that LLM safety operates as a sequential pipeline with distinct checkpoints. We introduce the \\textbf{Four-Checkpoint Framework}, which organizes safety mechanisms along two dimensions: processing stage (input vs.\\ output) and detection level (literal vs.\\ intent). This creates four chec","authors_text":"Hayfa Dhahbi, Kashyap Thimmaraju","cross_cats":["cs.AI","cs.CY","cs.ET","cs.HC"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2026-02-10T10:17:25Z","title":"Stop Testing Attacks, Start Diagnosing Defenses: The Four-Checkpoint Framework Reveals Where LLM Safety Breaks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.09629","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:ca18d50df0a61c55217b241d8fbee816f5c9b4f89104f0c1b267abe1e65ca6c8","target":"record","created_at":"2026-06-23T00:11:47Z","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":"94132c4198a4ee3f51ea0196430a577b53033a427e2e89a1260a85828bf9b8a1","cross_cats_sorted":["cs.AI","cs.CY","cs.ET","cs.HC"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2026-02-10T10:17:25Z","title_canon_sha256":"4c9a04e9b2f581009370ff62e16fc0802f3d6ae5f092455b665ec5cd29c858e4"},"schema_version":"1.0","source":{"id":"2602.09629","kind":"arxiv","version":1}},"canonical_sha256":"723408e8ec4b112c89f11b7f527b15aa61902ad49758b72c43c15387afaba5d3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"723408e8ec4b112c89f11b7f527b15aa61902ad49758b72c43c15387afaba5d3","first_computed_at":"2026-06-23T00:11:47.079050Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T00:11:47.079050Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rTxL2R9zM+7mPsd2A+2dIj45DdjI1aN52k5YhnZMeHYfAa3T/neR+FOAUaFcpmouS92SmCCSC3mpmKcucMnFCQ==","signature_status":"signed_v1","signed_at":"2026-06-23T00:11:47.079593Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.09629","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ca18d50df0a61c55217b241d8fbee816f5c9b4f89104f0c1b267abe1e65ca6c8","sha256:dee1bf7bde4a2b4870926d53684e1c9ca774b7be4d25b9c5032b28dbfef9061c"],"state_sha256":"e4d155ec6b79d4af454dbd37aef5646498a6b0efd9ec141d25fa2b6b27c8992e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/Uq17/8e/XP4AHkDU23J7k4eZiuX+B3XvBS7x6xh/UUhgcdQisZhudpBLQz5b0ex2wsMOONVfnJHEa9oAfa/DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-24T18:55:17.297902Z","bundle_sha256":"39ddbd887171dc7e98b1ff6bbcf1833f4d9a56bf9dfb96d6e4cf038ada12d3d7"}}