{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:HVL6X3YFBVHQ7BT7MNVUPMLIPB","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":"3fde68b965a33163d13b16e170ff32a27bfaac2613f8f6699d95e4902fa9e6fc","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2026-02-12T21:17:32Z","title_canon_sha256":"f9da5fe5a913d9fa36237ece97b4b63d4d513db81b2653805cba4fab69d2c472"},"schema_version":"1.0","source":{"id":"2602.12418","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.12418","created_at":"2026-06-30T01:16:28Z"},{"alias_kind":"arxiv_version","alias_value":"2602.12418v2","created_at":"2026-06-30T01:16:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.12418","created_at":"2026-06-30T01:16:28Z"},{"alias_kind":"pith_short_12","alias_value":"HVL6X3YFBVHQ","created_at":"2026-06-30T01:16:28Z"},{"alias_kind":"pith_short_16","alias_value":"HVL6X3YFBVHQ7BT7","created_at":"2026-06-30T01:16:28Z"},{"alias_kind":"pith_short_8","alias_value":"HVL6X3YF","created_at":"2026-06-30T01:16:28Z"}],"graph_snapshots":[{"event_id":"sha256:bdec238d966f9a2ea04259972c689ed9344b8c8d6bac4e53c0d8088a9db81854","target":"graph","created_at":"2026-06-30T01:16:28Z","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/2602.12418/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Jailbreak attacks remain a persistent threat to large language model safety. We propose Context-Conditioned Delta Steering (CC-Delta), an SAE-based defense that identifies jailbreak-relevant sparse features by comparing token-level representations of the same harmful request with and without jailbreak context. Using paired harmful/jailbreak prompts, CC-Delta selects features via statistical testing and applies inference-time mean-shift steering in SAE latent space. Across four aligned instruction-tuned models and twelve jailbreak attacks, CC-Delta achieves comparable or better safety-utility t","authors_text":"Arno Blaas, Jacopo Cortellazzi, Javier Abad, Pau Rodriguez, Xavier Suau, Yannick Assogba","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2026-02-12T21:17:32Z","title":"Sparse Autoencoders are Capable LLM Jailbreak Mitigators"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.12418","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":"sha256:9f7c57f601e0299d14bab183a5c3c029e9ec3dfb9f8acb54b84a29f8b5661393","target":"record","created_at":"2026-06-30T01:16:28Z","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":"3fde68b965a33163d13b16e170ff32a27bfaac2613f8f6699d95e4902fa9e6fc","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2026-02-12T21:17:32Z","title_canon_sha256":"f9da5fe5a913d9fa36237ece97b4b63d4d513db81b2653805cba4fab69d2c472"},"schema_version":"1.0","source":{"id":"2602.12418","kind":"arxiv","version":2}},"canonical_sha256":"3d57ebef050d4f0f867f636b47b168784bcb7b4a78ad491d1956edee13cd6672","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3d57ebef050d4f0f867f636b47b168784bcb7b4a78ad491d1956edee13cd6672","first_computed_at":"2026-06-30T01:16:28.304857Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-30T01:16:28.304857Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NIWuafrWNEu1/8WqDLPs52yYRvohrsUqzMeP40u6gQHZd0OhrKsH/Y+rn8eft7eTRWBLabh6Kb/D+V5ixr0mAg==","signature_status":"signed_v1","signed_at":"2026-06-30T01:16:28.305394Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.12418","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9f7c57f601e0299d14bab183a5c3c029e9ec3dfb9f8acb54b84a29f8b5661393","sha256:bdec238d966f9a2ea04259972c689ed9344b8c8d6bac4e53c0d8088a9db81854"],"state_sha256":"cc5fa266e400785d14aa2a9b5a3b2e79655eb2362a3187a3a06ff1464dc4a7d4"}