{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:5B7RV4TJQHTNC2EGAJENX3C4XO","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":"6e27de801ff9cb83b4f0c60a6b972aadc1b92eea6f8b45b4428b2f8356dbf29a","cross_cats_sorted":["cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-08T13:38:54Z","title_canon_sha256":"fd9fc2c42a6b4e4945c7d7ae87d5bd8542a4db0bd83a3c316f7fd4e32d29a80b"},"schema_version":"1.0","source":{"id":"2607.07405","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.07405","created_at":"2026-07-09T01:20:24Z"},{"alias_kind":"arxiv_version","alias_value":"2607.07405v1","created_at":"2026-07-09T01:20:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.07405","created_at":"2026-07-09T01:20:24Z"},{"alias_kind":"pith_short_12","alias_value":"5B7RV4TJQHTN","created_at":"2026-07-09T01:20:24Z"},{"alias_kind":"pith_short_16","alias_value":"5B7RV4TJQHTNC2EG","created_at":"2026-07-09T01:20:24Z"},{"alias_kind":"pith_short_8","alias_value":"5B7RV4TJ","created_at":"2026-07-09T01:20:24Z"}],"graph_snapshots":[{"event_id":"sha256:a294dd3d60ecfcd5c49f0911c710865dd1bc74696f7a69369b8d416b2c7e1137","target":"graph","created_at":"2026-07-09T01:20:24Z","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/2607.07405/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Tool-using LLM agents can violate the very policies they are deployed to enforce while appearing to complete the task successfully. In policy-permissive environments, a tool may execute any well-formed call even when the corresponding state transition is forbidden by domain policy. The result is a silent wrong state (a booking cancelled, a passenger count changed, a claim acted on without verification) that neither the tool nor the agent's self-report exposes.\n  We study this failure mode in the $\\tau^2$-bench airline domain. On a budget agent, 78% of observed failures are silent wrong-state f","authors_text":"Abhishek Basu, Sumanth Reddy Challaram, Vikas Reddy","cross_cats":["cs.CR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-08T13:38:54Z","title":"Reason Less, Verify More: Deterministic Gates Recover a Silent Policy-Violation Failure Mode in Tool-Using LLM Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.07405","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":"sha256:8c5d4546b4f785a91190654b36e07f4fe3009a62c3c0379a0e8daece111697a6","target":"record","created_at":"2026-07-09T01:20:24Z","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":"6e27de801ff9cb83b4f0c60a6b972aadc1b92eea6f8b45b4428b2f8356dbf29a","cross_cats_sorted":["cs.CR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-08T13:38:54Z","title_canon_sha256":"fd9fc2c42a6b4e4945c7d7ae87d5bd8542a4db0bd83a3c316f7fd4e32d29a80b"},"schema_version":"1.0","source":{"id":"2607.07405","kind":"arxiv","version":1}},"canonical_sha256":"e87f1af26981e6d168860248dbec5cbbbb94fa87b18e0dfca7d2d028f18e9109","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e87f1af26981e6d168860248dbec5cbbbb94fa87b18e0dfca7d2d028f18e9109","first_computed_at":"2026-07-09T01:20:24.121335Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-09T01:20:24.121335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dYwJz5tH06qEywOHhs5g/xLCeOonpjVj1lURoFBqailT7Isjn+p94nqubsawMpXsP1J5fiLEtRdcOm3IhNfHAw==","signature_status":"signed_v1","signed_at":"2026-07-09T01:20:24.121751Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.07405","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8c5d4546b4f785a91190654b36e07f4fe3009a62c3c0379a0e8daece111697a6","sha256:a294dd3d60ecfcd5c49f0911c710865dd1bc74696f7a69369b8d416b2c7e1137"],"state_sha256":"3f371ae452cba0fa49c6bac2fb3719bc7688f93b3224a33991b5ee7ab5a8cdda"}