{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:VMUJYAMUBWAIONZ3M6A6YCMCRN","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":"a3d5415ec1f596dd70774004da02a520df5f277db59d545169ca8ea417d118e3","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-07-12T20:37:19Z","title_canon_sha256":"6729766bf0b74202e8167b479ec4396b19dfa4bf692d4ae9956ce2590cfdda17"},"schema_version":"1.0","source":{"id":"2607.13075","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.13075","created_at":"2026-07-16T00:21:54Z"},{"alias_kind":"arxiv_version","alias_value":"2607.13075v1","created_at":"2026-07-16T00:21:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.13075","created_at":"2026-07-16T00:21:54Z"},{"alias_kind":"pith_short_12","alias_value":"VMUJYAMUBWAI","created_at":"2026-07-16T00:21:54Z"},{"alias_kind":"pith_short_16","alias_value":"VMUJYAMUBWAIONZ3","created_at":"2026-07-16T00:21:54Z"},{"alias_kind":"pith_short_8","alias_value":"VMUJYAMU","created_at":"2026-07-16T00:21:54Z"}],"graph_snapshots":[{"event_id":"sha256:2a636d5b3714312c62e2701eba05608527108a4ddb63c30546174b0b361de7c6","target":"graph","created_at":"2026-07-16T00:21:54Z","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/2607.13075/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Context can change whether a request is harmful without changing its topic or surface form. We ask whether residual-stream probes distinguish harmful requests from surface-matched benign controls at a useful operating point. Across three 7-8B model families, an activation sensor blocks 95.5-97.7 percent of judge-classified compliant attacks in a taxonomy-selected set. It also blocks 59.6-68.4 percent of XSTest prompts. A fully disjoint audit reconstructs near-ceiling source-contrast AUROC (0.996-0.999), but fixed transfer to matched pairs is weaker: 0.656-0.819 on the guard-selected Twin-n70 s","authors_text":"Dominik Schwarz","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-07-12T20:37:19Z","title":"The Entanglement Wall: Activation-Space Probes as Risk Detectors, Not Context Adjudicators"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.13075","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":"sha256:93287b342076afee976090269629c2c0098e2c9752a438b4c4fd43eba2ccc6a1","target":"record","created_at":"2026-07-16T00:21:54Z","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":"a3d5415ec1f596dd70774004da02a520df5f277db59d545169ca8ea417d118e3","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-07-12T20:37:19Z","title_canon_sha256":"6729766bf0b74202e8167b479ec4396b19dfa4bf692d4ae9956ce2590cfdda17"},"schema_version":"1.0","source":{"id":"2607.13075","kind":"arxiv","version":1}},"canonical_sha256":"ab289c01940d8087373b6781ec09828b6e87d8f7b556287c53a464acb8bffd25","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ab289c01940d8087373b6781ec09828b6e87d8f7b556287c53a464acb8bffd25","first_computed_at":"2026-07-16T00:21:54.590433Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-16T00:21:54.590433Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mBEOjqBBD3VXe1k9bOVUrsylnCWioiPKceh7BerfkU+It12UmbdG9IWh6lwF5+Sj2/63h3Lx0P0Z/+/1DF6JCA==","signature_status":"signed_v1","signed_at":"2026-07-16T00:21:54.591372Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.13075","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:93287b342076afee976090269629c2c0098e2c9752a438b4c4fd43eba2ccc6a1","sha256:2a636d5b3714312c62e2701eba05608527108a4ddb63c30546174b0b361de7c6"],"state_sha256":"fe6a8ba1efb9bba8565d4e14b6bcbd9d65867f204f2616f71f596f22dc298e71"}