{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:XA57BTIWPDAK4F6HKTZHIOGHPP","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":"93d3e69c2b987a7a05689edb830c8be475c87def4f7a3d8394762e55fb97ade5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-15T14:21:43Z","title_canon_sha256":"279880e78626c24b14cc657462e90e6ab557b628e8f8f2ac6e002d77cdcbd8ac"},"schema_version":"1.0","source":{"id":"2602.14161","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.14161","created_at":"2026-07-21T01:20:43Z"},{"alias_kind":"arxiv_version","alias_value":"2602.14161v2","created_at":"2026-07-21T01:20:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.14161","created_at":"2026-07-21T01:20:43Z"},{"alias_kind":"pith_short_12","alias_value":"XA57BTIWPDAK","created_at":"2026-07-21T01:20:43Z"},{"alias_kind":"pith_short_16","alias_value":"XA57BTIWPDAK4F6H","created_at":"2026-07-21T01:20:43Z"},{"alias_kind":"pith_short_8","alias_value":"XA57BTIW","created_at":"2026-07-21T01:20:43Z"}],"graph_snapshots":[{"event_id":"sha256:9b5fc3d45a508943c9696f6e0f5427c0aaf5223ff4507c2e084d947e12c9ab43","target":"graph","created_at":"2026-07-21T01:20:43Z","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/2602.14161/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Detecting prompt injection, jailbreak attacks, and harmful requests is critical for deploying LLM-based agents safely, yet current evaluation practices in this literature overestimate generalization. We train activation-based classifiers (linear probes on LLM hidden states) on a benchmark of 18 datasets (prompt attacks plus benign sources) and propose Leave-One-Dataset-Out (LODO) evaluation, where the held-out dataset is never seen during training. Across four LLMs from three families (Llama-3.1-8B, Gemma-3-27B, Qwen-3.5-2B/4B), standard cross-validation reports a pooled AUC 8.0-16.5 points hi","authors_text":"Max Fomin","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-15T14:21:43Z","title":"When Benchmarks Lie: Evaluating Malicious Prompt Classifiers Under True Distribution Shift"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.14161","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":"sha256:7dc626eb8c00725f750092fdb1046feb94b069dc259dacf958a93baf44ad701d","target":"record","created_at":"2026-07-21T01:20:43Z","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":"93d3e69c2b987a7a05689edb830c8be475c87def4f7a3d8394762e55fb97ade5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-15T14:21:43Z","title_canon_sha256":"279880e78626c24b14cc657462e90e6ab557b628e8f8f2ac6e002d77cdcbd8ac"},"schema_version":"1.0","source":{"id":"2602.14161","kind":"arxiv","version":2}},"canonical_sha256":"b83bf0cd1678c0ae17c754f27438c77bc1c313e4433924873ef4ca71fe149a2d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b83bf0cd1678c0ae17c754f27438c77bc1c313e4433924873ef4ca71fe149a2d","first_computed_at":"2026-07-21T01:20:43.987070Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-21T01:20:43.987070Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"X/y7Kwb5fgsw68lSc7y3XjCPZqI09ljkdpQTcMBizSITAK+NyOcqht01OcwLe85Euoy4q66nNgqIXilvT7r4BQ==","signature_status":"signed_v1","signed_at":"2026-07-21T01:20:43.988144Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.14161","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7dc626eb8c00725f750092fdb1046feb94b069dc259dacf958a93baf44ad701d","sha256:9b5fc3d45a508943c9696f6e0f5427c0aaf5223ff4507c2e084d947e12c9ab43"],"state_sha256":"bfef5ca4a286635ed66e3b9bf19b4d1d08ad776724b6d2d8803cde6d7845680e"}