{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:2RCNDXAD3Z5USTOMNOSUIE35BY","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":"a8bb9982e9d0b77876620d85a4227af7992e0620d18640110f5fc506fd392bdf","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-10-25T19:18:22Z","title_canon_sha256":"a53e893ef6c824f586206a65dd01f0fa69865b719f4ea70da193553400794b2e"},"schema_version":"1.0","source":{"id":"2410.19937","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.19937","created_at":"2026-07-05T09:26:35Z"},{"alias_kind":"arxiv_version","alias_value":"2410.19937v1","created_at":"2026-07-05T09:26:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.19937","created_at":"2026-07-05T09:26:35Z"},{"alias_kind":"pith_short_12","alias_value":"2RCNDXAD3Z5U","created_at":"2026-07-05T09:26:35Z"},{"alias_kind":"pith_short_16","alias_value":"2RCNDXAD3Z5USTOM","created_at":"2026-07-05T09:26:35Z"},{"alias_kind":"pith_short_8","alias_value":"2RCNDXAD","created_at":"2026-07-05T09:26:35Z"}],"graph_snapshots":[{"event_id":"sha256:5765477f7decc0603a4502f4e376a5fa274a210a654e84ea2d1455f92fcb6990","target":"graph","created_at":"2026-07-05T09:26:35Z","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/2410.19937/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Jailbreak attacks circumvent LLMs' built-in safeguards by concealing harmful queries within jailbreak prompts. While existing defenses primarily focus on mitigating the effects of jailbreak prompts, they often prove inadequate as jailbreak prompts can take arbitrary, adaptive forms. This paper presents RobustKV, a novel defense that adopts a fundamentally different approach by selectively removing critical tokens of harmful queries from key-value (KV) caches. Intuitively, for a jailbreak prompt to be effective, its tokens must achieve sufficient `importance' (as measured by attention scores), ","authors_text":"Changjiang Li, Jiacheng Liang, Tanqiu Jiang, Ting Wang, Yuhui Wang, Zian Wang","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-10-25T19:18:22Z","title":"RobustKV: Defending Large Language Models against Jailbreak Attacks via KV Eviction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.19937","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":"sha256:2ff862cd39177314aabdd6849ea061da60c23d7c77495e23a71c9108154c17e6","target":"record","created_at":"2026-07-05T09:26:35Z","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":"a8bb9982e9d0b77876620d85a4227af7992e0620d18640110f5fc506fd392bdf","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2024-10-25T19:18:22Z","title_canon_sha256":"a53e893ef6c824f586206a65dd01f0fa69865b719f4ea70da193553400794b2e"},"schema_version":"1.0","source":{"id":"2410.19937","kind":"arxiv","version":1}},"canonical_sha256":"d444d1dc03de7b494dcc6ba544137d0e02810275a29389da6f255ce5b538c4a0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d444d1dc03de7b494dcc6ba544137d0e02810275a29389da6f255ce5b538c4a0","first_computed_at":"2026-07-05T09:26:35.913482Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:26:35.913482Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"d54/v0lNTfIFvtmPBQDbW4YgF5iQRgmNgGdj+aMZNH0IrCfx2eK9W/cNele0lo7i4QHHMkRVhGKPodek9PrGBg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:26:35.913966Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.19937","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2ff862cd39177314aabdd6849ea061da60c23d7c77495e23a71c9108154c17e6","sha256:5765477f7decc0603a4502f4e376a5fa274a210a654e84ea2d1455f92fcb6990"],"state_sha256":"b33fa1bbff0953fc8c2d5cd552f5a6667f1a1dee6263c5b010eb44df690c6267"}