{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:AJATFAUB6272JOVHX2OK2THYKL","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":"a29459e2b6f978952a9904fd3250c2e86b0f8e997fd25b11f4d8bfb58cf739e1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-07-08T20:31:06Z","title_canon_sha256":"8e5a1a8402ea7a6daf3d0292baaa3182e3a010c5a76b7808c8b22fa6dd844eab"},"schema_version":"1.0","source":{"id":"2607.07903","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.07903","created_at":"2026-07-10T00:19:04Z"},{"alias_kind":"arxiv_version","alias_value":"2607.07903v1","created_at":"2026-07-10T00:19:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.07903","created_at":"2026-07-10T00:19:04Z"},{"alias_kind":"pith_short_12","alias_value":"AJATFAUB6272","created_at":"2026-07-10T00:19:04Z"},{"alias_kind":"pith_short_16","alias_value":"AJATFAUB6272JOVH","created_at":"2026-07-10T00:19:04Z"},{"alias_kind":"pith_short_8","alias_value":"AJATFAUB","created_at":"2026-07-10T00:19:04Z"}],"graph_snapshots":[{"event_id":"sha256:2f7d7c889871bba6065c64b6efdbe1ba4803d372cba7d8c53cbd567f7e997218","target":"graph","created_at":"2026-07-10T00:19:04Z","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/2607.07903/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) exhibit remarkable capabilities but remain highly vulnerable to adversarial prompts and jailbreak attacks. Existing approaches primarily analyze these failures through input-output behaviors or attribution methods, offering limited insight into how adversarial perturbations alter the model's internal reasoning. Consequently, the mechanisms underlying unsafe or incorrect behaviors remain poorly understood. We introduce a mechanistic framework for diagnosing LLM vulnerabilities using paired internal computation graphs, which represent prompt-specific inference as str","authors_text":"Anupam Wagle, Chaowei Zhang, Ifrat Ikhtear Uddin, Longwei Wang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-07-08T20:31:06Z","title":"Mechanistic Interpretability of LLM Jailbreaks via Internal Attribution Graphs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.07903","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:b71b5c63499be89c5cb6aaf1c338dde750dc3cd8305c01e44d0011fb22492991","target":"record","created_at":"2026-07-10T00:19:04Z","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":"a29459e2b6f978952a9904fd3250c2e86b0f8e997fd25b11f4d8bfb58cf739e1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-07-08T20:31:06Z","title_canon_sha256":"8e5a1a8402ea7a6daf3d0292baaa3182e3a010c5a76b7808c8b22fa6dd844eab"},"schema_version":"1.0","source":{"id":"2607.07903","kind":"arxiv","version":1}},"canonical_sha256":"0241328281f6bfa4baa7be9cad4cf852c41025ccc9561a730b58ea0d2eb15183","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0241328281f6bfa4baa7be9cad4cf852c41025ccc9561a730b58ea0d2eb15183","first_computed_at":"2026-07-10T00:19:04.447633Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-10T00:19:04.447633Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uUxUgRggjioF+Zi+YX3pSt4Sw1HQQdCZE84/UPOUmx0p0dsxE618rY/obbbmrbCB4V6t/D36ucdrjHkuoB8UBw==","signature_status":"signed_v1","signed_at":"2026-07-10T00:19:04.448030Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.07903","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b71b5c63499be89c5cb6aaf1c338dde750dc3cd8305c01e44d0011fb22492991","sha256:2f7d7c889871bba6065c64b6efdbe1ba4803d372cba7d8c53cbd567f7e997218"],"state_sha256":"04e677e0fc5db011227119f7e024976c79f59e084c801e0a4b80d33a06309165"}