{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:4EIWVIGMJ3WM36EBMYNUAHR4QM","short_pith_number":"pith:4EIWVIGM","canonical_record":{"source":{"id":"2502.13162","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2025-02-16T18:47:41Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"36989baaac9ff08ce2eb0de2a38ee3d7569b584618176cd94af85fd59dc54169","abstract_canon_sha256":"5092d030ff1eaba4c3bf017485d520b54dfd5a3fd98803ea21f466eb8d37f9f4"},"schema_version":"1.0"},"canonical_sha256":"e1116aa0cc4eeccdf881661b401e3c8328522ee5331dd076385c1b69f42c8ae2","source":{"kind":"arxiv","id":"2502.13162","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.13162","created_at":"2026-07-05T10:16:43Z"},{"alias_kind":"arxiv_version","alias_value":"2502.13162v1","created_at":"2026-07-05T10:16:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.13162","created_at":"2026-07-05T10:16:43Z"},{"alias_kind":"pith_short_12","alias_value":"4EIWVIGMJ3WM","created_at":"2026-07-05T10:16:43Z"},{"alias_kind":"pith_short_16","alias_value":"4EIWVIGMJ3WM36EB","created_at":"2026-07-05T10:16:43Z"},{"alias_kind":"pith_short_8","alias_value":"4EIWVIGM","created_at":"2026-07-05T10:16:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:4EIWVIGMJ3WM36EBMYNUAHR4QM","target":"record","payload":{"canonical_record":{"source":{"id":"2502.13162","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2025-02-16T18:47:41Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"36989baaac9ff08ce2eb0de2a38ee3d7569b584618176cd94af85fd59dc54169","abstract_canon_sha256":"5092d030ff1eaba4c3bf017485d520b54dfd5a3fd98803ea21f466eb8d37f9f4"},"schema_version":"1.0"},"canonical_sha256":"e1116aa0cc4eeccdf881661b401e3c8328522ee5331dd076385c1b69f42c8ae2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:16:43.273449Z","signature_b64":"fgHgHmpH+/7aJuGaiLBmsNBIigtxZIsPk3/IYyMOZcFdHpH8OJLZnXwROwLlzm4HtPKXIN9MuTCH1nidpXXLAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e1116aa0cc4eeccdf881661b401e3c8328522ee5331dd076385c1b69f42c8ae2","last_reissued_at":"2026-07-05T10:16:43.272962Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:16:43.272962Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.13162","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":"2026-07-05T10:16:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TcRTYpmY9JiKj3oKCAxAvunml7phfLnPAZPFipDPJLJu1pQBLzfeZWDGDRmDE+irD8HbEkmUYkL+B5rjSsVXCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T15:15:57.788718Z"},"content_sha256":"464e4f1371249fc5aa925ec59e67921584d1d519615d7398e03c09c6f0314c28","schema_version":"1.0","event_id":"sha256:464e4f1371249fc5aa925ec59e67921584d1d519615d7398e03c09c6f0314c28"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:4EIWVIGMJ3WM36EBMYNUAHR4QM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ShieldLearner: A New Paradigm for Jailbreak Attack Defense in LLMs","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CR","authors_text":"Hao Wang, Huacan Wang, Ziyi Ni","submitted_at":"2025-02-16T18:47:41Z","abstract_excerpt":"Large Language Models (LLMs) have achieved remarkable success in various domains but remain vulnerable to adversarial jailbreak attacks. Existing prompt-defense strategies, including parameter-modifying and parameter-free approaches, face limitations in adaptability, interpretability, and customization, constraining their effectiveness against evolving threats. To address these challenges, we propose ShieldLearner, a novel paradigm that mimics human learning in defense. Through trial and error, it autonomously distills attack signatures into a Pattern Atlas and synthesizes defense heuristics i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.13162","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/2502.13162/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":"2026-07-05T10:16:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HxeQz8t+csGD/MSUW8dOf1fbt4NtWY8tsIJwE71mjS6bHJVYVR5Y58+05WwwlmfKA/Cxzv3dMorvU/L/8kt2Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T15:15:57.789648Z"},"content_sha256":"51737cbc3af34d286ad5afab8027a864b8afb8a08f1df47355f34f7985372291","schema_version":"1.0","event_id":"sha256:51737cbc3af34d286ad5afab8027a864b8afb8a08f1df47355f34f7985372291"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4EIWVIGMJ3WM36EBMYNUAHR4QM/bundle.json","state_url":"https://pith.science/pith/4EIWVIGMJ3WM36EBMYNUAHR4QM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4EIWVIGMJ3WM36EBMYNUAHR4QM/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":"2026-08-04T15:15:57Z","links":{"resolver":"https://pith.science/pith/4EIWVIGMJ3WM36EBMYNUAHR4QM","bundle":"https://pith.science/pith/4EIWVIGMJ3WM36EBMYNUAHR4QM/bundle.json","state":"https://pith.science/pith/4EIWVIGMJ3WM36EBMYNUAHR4QM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4EIWVIGMJ3WM36EBMYNUAHR4QM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4EIWVIGMJ3WM36EBMYNUAHR4QM","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":"5092d030ff1eaba4c3bf017485d520b54dfd5a3fd98803ea21f466eb8d37f9f4","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2025-02-16T18:47:41Z","title_canon_sha256":"36989baaac9ff08ce2eb0de2a38ee3d7569b584618176cd94af85fd59dc54169"},"schema_version":"1.0","source":{"id":"2502.13162","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.13162","created_at":"2026-07-05T10:16:43Z"},{"alias_kind":"arxiv_version","alias_value":"2502.13162v1","created_at":"2026-07-05T10:16:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.13162","created_at":"2026-07-05T10:16:43Z"},{"alias_kind":"pith_short_12","alias_value":"4EIWVIGMJ3WM","created_at":"2026-07-05T10:16:43Z"},{"alias_kind":"pith_short_16","alias_value":"4EIWVIGMJ3WM36EB","created_at":"2026-07-05T10:16:43Z"},{"alias_kind":"pith_short_8","alias_value":"4EIWVIGM","created_at":"2026-07-05T10:16:43Z"}],"graph_snapshots":[{"event_id":"sha256:51737cbc3af34d286ad5afab8027a864b8afb8a08f1df47355f34f7985372291","target":"graph","created_at":"2026-07-05T10:16:43Z","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/2502.13162/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have achieved remarkable success in various domains but remain vulnerable to adversarial jailbreak attacks. Existing prompt-defense strategies, including parameter-modifying and parameter-free approaches, face limitations in adaptability, interpretability, and customization, constraining their effectiveness against evolving threats. To address these challenges, we propose ShieldLearner, a novel paradigm that mimics human learning in defense. Through trial and error, it autonomously distills attack signatures into a Pattern Atlas and synthesizes defense heuristics i","authors_text":"Hao Wang, Huacan Wang, Ziyi Ni","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2025-02-16T18:47:41Z","title":"ShieldLearner: A New Paradigm for Jailbreak Attack Defense in LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.13162","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":"sha256:464e4f1371249fc5aa925ec59e67921584d1d519615d7398e03c09c6f0314c28","target":"record","created_at":"2026-07-05T10:16:43Z","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":"5092d030ff1eaba4c3bf017485d520b54dfd5a3fd98803ea21f466eb8d37f9f4","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2025-02-16T18:47:41Z","title_canon_sha256":"36989baaac9ff08ce2eb0de2a38ee3d7569b584618176cd94af85fd59dc54169"},"schema_version":"1.0","source":{"id":"2502.13162","kind":"arxiv","version":1}},"canonical_sha256":"e1116aa0cc4eeccdf881661b401e3c8328522ee5331dd076385c1b69f42c8ae2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e1116aa0cc4eeccdf881661b401e3c8328522ee5331dd076385c1b69f42c8ae2","first_computed_at":"2026-07-05T10:16:43.272962Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:16:43.272962Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fgHgHmpH+/7aJuGaiLBmsNBIigtxZIsPk3/IYyMOZcFdHpH8OJLZnXwROwLlzm4HtPKXIN9MuTCH1nidpXXLAA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:16:43.273449Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.13162","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:464e4f1371249fc5aa925ec59e67921584d1d519615d7398e03c09c6f0314c28","sha256:51737cbc3af34d286ad5afab8027a864b8afb8a08f1df47355f34f7985372291"],"state_sha256":"b40cf7951fea35e1244c82be717470a94e4c4357d286d49f0788310a425b0dba"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wocZ26xU/23Ti73nxhSSu3t6XurweMosQ5i7FZoY9B30rWjOc7gR7wtUbzSuSk0ogoHMubFProRK7ipU6jOqDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T15:15:57.795002Z","bundle_sha256":"9884992ace7a51d4e8b4fdae010dc2ae32105977a2744ef8bb1bc1611c9678f7"}}