{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:HAQCIVTZN2W5QQ6CSJCRL6NHYD","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":"95873eb13d04c4ca6234b1b2430047707b6dd8c8aae9c1e4ebccb845ed4b16b4","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-18T09:02:12Z","title_canon_sha256":"39189eb990b2a012eb1f81ba5883979dc41840457695ea8998c7e939a9a35c6d"},"schema_version":"1.0","source":{"id":"2504.13562","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.13562","created_at":"2026-07-05T10:50:57Z"},{"alias_kind":"arxiv_version","alias_value":"2504.13562v1","created_at":"2026-07-05T10:50:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.13562","created_at":"2026-07-05T10:50:57Z"},{"alias_kind":"pith_short_12","alias_value":"HAQCIVTZN2W5","created_at":"2026-07-05T10:50:57Z"},{"alias_kind":"pith_short_16","alias_value":"HAQCIVTZN2W5QQ6C","created_at":"2026-07-05T10:50:57Z"},{"alias_kind":"pith_short_8","alias_value":"HAQCIVTZ","created_at":"2026-07-05T10:50:57Z"}],"graph_snapshots":[{"event_id":"sha256:823991ae96742ca2cac3a92750d278efdc70095296bc919ebb58209391fd2060","target":"graph","created_at":"2026-07-05T10:50:57Z","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/2504.13562/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the widespread adoption of Large Language Models (LLMs), jailbreak attacks have become an increasingly pressing safety concern. While safety-aligned LLMs can effectively defend against normal harmful queries, they remain vulnerable to such attacks. Existing defense methods primarily rely on fine-tuning or input modification, which often suffer from limited generalization and reduced utility. To address this, we introduce DETAM, a finetuning-free defense approach that improves the defensive capabilities against jailbreak attacks of LLMs via targeted attention modification. Specifically, we","authors_text":"Han Jiang, Yu Li, Zhihua Wei","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-18T09:02:12Z","title":"DETAM: Defending LLMs Against Jailbreak Attacks via Targeted Attention Modification"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.13562","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":"sha256:158c51d75cc22e363e2d2c8f8b4725d46fa82bf0f0c8f200ff6572b49589825d","target":"record","created_at":"2026-07-05T10:50:57Z","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":"95873eb13d04c4ca6234b1b2430047707b6dd8c8aae9c1e4ebccb845ed4b16b4","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-04-18T09:02:12Z","title_canon_sha256":"39189eb990b2a012eb1f81ba5883979dc41840457695ea8998c7e939a9a35c6d"},"schema_version":"1.0","source":{"id":"2504.13562","kind":"arxiv","version":1}},"canonical_sha256":"38202456796eadd843c2924515f9a7c0c5d206751dcd796ca4bb197a2747a6b3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"38202456796eadd843c2924515f9a7c0c5d206751dcd796ca4bb197a2747a6b3","first_computed_at":"2026-07-05T10:50:57.949769Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:50:57.949769Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RVqYWXwklmSjhMmR7bmT25+ejk81GFN6KYcdiQtB1y7ixqMLBjdGVK7SynokBsfvjZ8Eo0mh36Hp9NAm6VaXBA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:50:57.950391Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.13562","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:158c51d75cc22e363e2d2c8f8b4725d46fa82bf0f0c8f200ff6572b49589825d","sha256:823991ae96742ca2cac3a92750d278efdc70095296bc919ebb58209391fd2060"],"state_sha256":"d97388ddc3e0c1daa2ef272a720269add0a4fb2beb02844d99421bbfa5369c49"}