{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:WGB4VCPTET3OZP5WDVWFSOFGAF","short_pith_number":"pith:WGB4VCPT","canonical_record":{"source":{"id":"2406.18118","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-06-26T07:15:44Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"bcf9fc76a8a6c6e62ddd02e251b3c6499ad1c542a105e618777d5ee79b491ff3","abstract_canon_sha256":"c0d972ae123fa9eb0e7396398efafaed405aa96505b1e4d2c23a3908efe94799"},"schema_version":"1.0"},"canonical_sha256":"b183ca89f324f6ecbfb61d6c5938a6016335c8e93617c6ecab44abb39f8efb5b","source":{"kind":"arxiv","id":"2406.18118","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.18118","created_at":"2026-07-05T09:53:41Z"},{"alias_kind":"arxiv_version","alias_value":"2406.18118v4","created_at":"2026-07-05T09:53:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.18118","created_at":"2026-07-05T09:53:41Z"},{"alias_kind":"pith_short_12","alias_value":"WGB4VCPTET3O","created_at":"2026-07-05T09:53:41Z"},{"alias_kind":"pith_short_16","alias_value":"WGB4VCPTET3OZP5W","created_at":"2026-07-05T09:53:41Z"},{"alias_kind":"pith_short_8","alias_value":"WGB4VCPT","created_at":"2026-07-05T09:53:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:WGB4VCPTET3OZP5WDVWFSOFGAF","target":"record","payload":{"canonical_record":{"source":{"id":"2406.18118","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-06-26T07:15:44Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"bcf9fc76a8a6c6e62ddd02e251b3c6499ad1c542a105e618777d5ee79b491ff3","abstract_canon_sha256":"c0d972ae123fa9eb0e7396398efafaed405aa96505b1e4d2c23a3908efe94799"},"schema_version":"1.0"},"canonical_sha256":"b183ca89f324f6ecbfb61d6c5938a6016335c8e93617c6ecab44abb39f8efb5b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:53:41.993670Z","signature_b64":"Cq1tCqa7GNukinr+XKrO4fLXjaqvQMYFWUk4DQM7HK8N21Crssh2TwlbJUANd026WXULLmS39fRJF8GlA4m0AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b183ca89f324f6ecbfb61d6c5938a6016335c8e93617c6ecab44abb39f8efb5b","last_reissued_at":"2026-07-05T09:53:41.993178Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:53:41.993178Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.18118","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":"2026-07-05T09:53:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h7ljfL5CXDhhuw8eLHuzJhaim/FpAQgaTe4OkNkPGRAV55egr5K9LP8oNn8m5rv62rs6ENzfsUMXKo6Q8Ar2AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T00:14:02.187833Z"},"content_sha256":"350cb7250c354997f4ca3297621d854a9fabdace8878585ecbd3dd7f2ae21cf2","schema_version":"1.0","event_id":"sha256:350cb7250c354997f4ca3297621d854a9fabdace8878585ecbd3dd7f2ae21cf2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:WGB4VCPTET3OZP5WDVWFSOFGAF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SafeAligner: Safety Alignment against Jailbreak Attacks via Response Disparity Guidance","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CR","authors_text":"Caishuang Huang, Enyu Zhou, Huijie Lv, Junjie Ye, Qi Zhang, Rui Zheng, Shihan Dou, Sixian Li, Tao Gui, Wanxu Zhao, Wenyu Zhan, Xiao Wang, Xuanjing Huang, Yuming Yang","submitted_at":"2024-06-26T07:15:44Z","abstract_excerpt":"As the development of large language models (LLMs) rapidly advances, securing these models effectively without compromising their utility has become a pivotal area of research. However, current defense strategies against jailbreak attacks (i.e., efforts to bypass security protocols) often suffer from limited adaptability, restricted general capability, and high cost. To address these challenges, we introduce SafeAligner, a methodology implemented at the decoding stage to fortify defenses against jailbreak attacks. We begin by developing two specialized models: the Sentinel Model, which is trai"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.18118","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/2406.18118/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":"2026-07-05T09:53:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y4tjZr9ydNBCJMJzMucSwbaYXmeJVLpXKCHlX2URQxWVxPafQJPtWknoUGphD8rIlXoi/Os8IWZm+rHXvFX7Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T00:14:02.188366Z"},"content_sha256":"91a5207890d3527c7d7158242f0fad8e10cdc31e9ebca72b76e701d09abf5645","schema_version":"1.0","event_id":"sha256:91a5207890d3527c7d7158242f0fad8e10cdc31e9ebca72b76e701d09abf5645"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WGB4VCPTET3OZP5WDVWFSOFGAF/bundle.json","state_url":"https://pith.science/pith/WGB4VCPTET3OZP5WDVWFSOFGAF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WGB4VCPTET3OZP5WDVWFSOFGAF/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":"2026-08-11T00:14:02Z","links":{"resolver":"https://pith.science/pith/WGB4VCPTET3OZP5WDVWFSOFGAF","bundle":"https://pith.science/pith/WGB4VCPTET3OZP5WDVWFSOFGAF/bundle.json","state":"https://pith.science/pith/WGB4VCPTET3OZP5WDVWFSOFGAF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WGB4VCPTET3OZP5WDVWFSOFGAF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:WGB4VCPTET3OZP5WDVWFSOFGAF","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":"c0d972ae123fa9eb0e7396398efafaed405aa96505b1e4d2c23a3908efe94799","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-06-26T07:15:44Z","title_canon_sha256":"bcf9fc76a8a6c6e62ddd02e251b3c6499ad1c542a105e618777d5ee79b491ff3"},"schema_version":"1.0","source":{"id":"2406.18118","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.18118","created_at":"2026-07-05T09:53:41Z"},{"alias_kind":"arxiv_version","alias_value":"2406.18118v4","created_at":"2026-07-05T09:53:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.18118","created_at":"2026-07-05T09:53:41Z"},{"alias_kind":"pith_short_12","alias_value":"WGB4VCPTET3O","created_at":"2026-07-05T09:53:41Z"},{"alias_kind":"pith_short_16","alias_value":"WGB4VCPTET3OZP5W","created_at":"2026-07-05T09:53:41Z"},{"alias_kind":"pith_short_8","alias_value":"WGB4VCPT","created_at":"2026-07-05T09:53:41Z"}],"graph_snapshots":[{"event_id":"sha256:91a5207890d3527c7d7158242f0fad8e10cdc31e9ebca72b76e701d09abf5645","target":"graph","created_at":"2026-07-05T09:53:41Z","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/2406.18118/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As the development of large language models (LLMs) rapidly advances, securing these models effectively without compromising their utility has become a pivotal area of research. However, current defense strategies against jailbreak attacks (i.e., efforts to bypass security protocols) often suffer from limited adaptability, restricted general capability, and high cost. To address these challenges, we introduce SafeAligner, a methodology implemented at the decoding stage to fortify defenses against jailbreak attacks. We begin by developing two specialized models: the Sentinel Model, which is trai","authors_text":"Caishuang Huang, Enyu Zhou, Huijie Lv, Junjie Ye, Qi Zhang, Rui Zheng, Shihan Dou, Sixian Li, Tao Gui, Wanxu Zhao, Wenyu Zhan, Xiao Wang, Xuanjing Huang, Yuming Yang","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-06-26T07:15:44Z","title":"SafeAligner: Safety Alignment against Jailbreak Attacks via Response Disparity Guidance"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.18118","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":"sha256:350cb7250c354997f4ca3297621d854a9fabdace8878585ecbd3dd7f2ae21cf2","target":"record","created_at":"2026-07-05T09:53:41Z","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":"c0d972ae123fa9eb0e7396398efafaed405aa96505b1e4d2c23a3908efe94799","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-06-26T07:15:44Z","title_canon_sha256":"bcf9fc76a8a6c6e62ddd02e251b3c6499ad1c542a105e618777d5ee79b491ff3"},"schema_version":"1.0","source":{"id":"2406.18118","kind":"arxiv","version":4}},"canonical_sha256":"b183ca89f324f6ecbfb61d6c5938a6016335c8e93617c6ecab44abb39f8efb5b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b183ca89f324f6ecbfb61d6c5938a6016335c8e93617c6ecab44abb39f8efb5b","first_computed_at":"2026-07-05T09:53:41.993178Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:53:41.993178Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Cq1tCqa7GNukinr+XKrO4fLXjaqvQMYFWUk4DQM7HK8N21Crssh2TwlbJUANd026WXULLmS39fRJF8GlA4m0AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:53:41.993670Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.18118","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:350cb7250c354997f4ca3297621d854a9fabdace8878585ecbd3dd7f2ae21cf2","sha256:91a5207890d3527c7d7158242f0fad8e10cdc31e9ebca72b76e701d09abf5645"],"state_sha256":"1bb47a474a35b77b237fdec1bfbd1aacc5997551471ae56d6c701ff1452df82b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DQ/iBSm+aF0Em/f8/xVblJLq/jyXgo7p7O/8HPGYjEHjk/1jL53CCgErOd7RUi5lZSAoAk7ck+2kiZQChmZUCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T00:14:02.193780Z","bundle_sha256":"c3f41ba7659bc21b0ac2d67571d4b82933a4e77f6ea971927e51b4dafba43a67"}}