{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:P6SH75ZDTSN2K7BGOPU3OSDVVE","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":"ae883960b29417f2081a9e3376721e940205641a6089480cc7c422197f39316e","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-02-26T10:53:58Z","title_canon_sha256":"7cff4d4ea1f123f56c3d67ffb1e9338544199fc72ad6bbde26134c1942674af1"},"schema_version":"1.0","source":{"id":"2502.19041","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.19041","created_at":"2026-07-05T11:10:53Z"},{"alias_kind":"arxiv_version","alias_value":"2502.19041v2","created_at":"2026-07-05T11:10:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.19041","created_at":"2026-07-05T11:10:53Z"},{"alias_kind":"pith_short_12","alias_value":"P6SH75ZDTSN2","created_at":"2026-07-05T11:10:53Z"},{"alias_kind":"pith_short_16","alias_value":"P6SH75ZDTSN2K7BG","created_at":"2026-07-05T11:10:53Z"},{"alias_kind":"pith_short_8","alias_value":"P6SH75ZD","created_at":"2026-07-05T11:10:53Z"}],"graph_snapshots":[{"event_id":"sha256:f046d74c2d146406d0843cc071d3d2459f570ff9accf644ca34e7f75bae7fb33","target":"graph","created_at":"2026-07-05T11:10:53Z","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/2502.19041/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Although Aligned Large Language Models (LLMs) are trained to refuse harmful requests, they remain vulnerable to jailbreak attacks. Unfortunately, existing methods often focus on surface-level patterns, overlooking the deeper attack essences. As a result, defenses fail when attack prompts change, even though the underlying \"attack essence\" remains the same. To address this issue, we introduce EDDF, an \\textbf{E}ssence-\\textbf{D}riven \\textbf{D}efense \\textbf{F}ramework Against Jailbreak Attacks in LLMs. EDDF is a plug-and-play input-filtering method and operates in two stages: 1) offline essenc","authors_text":"Ansen Zhang, Ronghao Chen, Shiyu Xiang, Yanfei Cao, Yang Fan","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-02-26T10:53:58Z","title":"Beyond Surface-Level Patterns: An Essence-Driven Defense Framework Against Jailbreak Attacks in LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.19041","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:db37a8f9881241303ffb3945e960e9081e0ed8213626613bf51a6ba4ea85872a","target":"record","created_at":"2026-07-05T11:10:53Z","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":"ae883960b29417f2081a9e3376721e940205641a6089480cc7c422197f39316e","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-02-26T10:53:58Z","title_canon_sha256":"7cff4d4ea1f123f56c3d67ffb1e9338544199fc72ad6bbde26134c1942674af1"},"schema_version":"1.0","source":{"id":"2502.19041","kind":"arxiv","version":2}},"canonical_sha256":"7fa47ff7239c9ba57c2673e9b74875a932d70e572f3d45c1071ec86ada4271ab","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7fa47ff7239c9ba57c2673e9b74875a932d70e572f3d45c1071ec86ada4271ab","first_computed_at":"2026-07-05T11:10:53.465169Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:10:53.465169Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bBHMeOg1WPb9G+c/VaA9GKcoavc/k02RzSYUEkN9GgAlOC3tdoQ48awJtXrjv5S0/JKKsdsxFtxdImK3JJIMDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:10:53.465731Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.19041","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:db37a8f9881241303ffb3945e960e9081e0ed8213626613bf51a6ba4ea85872a","sha256:f046d74c2d146406d0843cc071d3d2459f570ff9accf644ca34e7f75bae7fb33"],"state_sha256":"f4059ad6d3396c9ffed4db45dfc23467fbd6912b6114dcce2c03d2562d4a646f"}