{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:I5EKPE2X3FYI5PIGVMY7Q5MP6N","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":"88f47476b5e93c3cea768ca22d33a0babf9e08834ea54ea9a240d2f3a74199a3","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-11T12:01:09Z","title_canon_sha256":"2a8bdadc39915e7590169a4ddacb8dbbe0f2e6229b7460367949df0da79640a1"},"schema_version":"1.0","source":{"id":"2406.07188","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.07188","created_at":"2026-07-05T08:43:28Z"},{"alias_kind":"arxiv_version","alias_value":"2406.07188v2","created_at":"2026-07-05T08:43:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.07188","created_at":"2026-07-05T08:43:28Z"},{"alias_kind":"pith_short_12","alias_value":"I5EKPE2X3FYI","created_at":"2026-07-05T08:43:28Z"},{"alias_kind":"pith_short_16","alias_value":"I5EKPE2X3FYI5PIG","created_at":"2026-07-05T08:43:28Z"},{"alias_kind":"pith_short_8","alias_value":"I5EKPE2X","created_at":"2026-07-05T08:43:28Z"}],"graph_snapshots":[{"event_id":"sha256:324b112062d3e2162188b46e4248c86b7b5bf5e9275f26b854c960dc36985e8f","target":"graph","created_at":"2026-07-05T08:43:28Z","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/2406.07188/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The robustness of large language models (LLMs) against adversarial manipulations, such as jailbreak attacks, remains a significant challenge. In this work, we propose an approach that enhances the self-critique capability of the LLM and further fine-tunes it over sanitized synthetic data. This is done with the addition of an external critic model that can be merged with the original, thus bolstering self-critique capabilities and improving the robustness of the LLMs response to adversarial prompts. Our results demonstrate that the combination of merging and self-critique can reduce the attack ","authors_text":"Victor Gallego","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-11T12:01:09Z","title":"Merging Improves Self-Critique Against Jailbreak Attacks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.07188","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":"sha256:1a40d4ae371bee553be921fe770ed37154673c0892561344fa8934543b255f06","target":"record","created_at":"2026-07-05T08:43:28Z","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":"88f47476b5e93c3cea768ca22d33a0babf9e08834ea54ea9a240d2f3a74199a3","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-11T12:01:09Z","title_canon_sha256":"2a8bdadc39915e7590169a4ddacb8dbbe0f2e6229b7460367949df0da79640a1"},"schema_version":"1.0","source":{"id":"2406.07188","kind":"arxiv","version":2}},"canonical_sha256":"4748a79357d9708ebd06ab31f8758ff360d5e17bb0c1c1cb16259bad08716fe5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4748a79357d9708ebd06ab31f8758ff360d5e17bb0c1c1cb16259bad08716fe5","first_computed_at":"2026-07-05T08:43:28.943250Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:43:28.943250Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8C72Ym2oY2jij5WlJJg0NjrBh9HS/KR4ZBLOmROExWHpKMdQI7NltChqMqRYw7hsjtD2mJA6R+KK7GEIuWIrCg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:43:28.943674Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.07188","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1a40d4ae371bee553be921fe770ed37154673c0892561344fa8934543b255f06","sha256:324b112062d3e2162188b46e4248c86b7b5bf5e9275f26b854c960dc36985e8f"],"state_sha256":"9a3f6c4977e0f8184d378a3f437fb86272f386397ff4a85245f55313920fc02e"}