{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:URQGX36RMICXEZ7RTMITJVZ2OV","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":"e25c8c3d37918987ec835d8cad138fa9629a1e66f593cdfe2ae583f52a9f8aff","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-27T15:39:46Z","title_canon_sha256":"48cdc9f8a2db77d92ba3e746711c7231cdb0304bcc2a2201c8b3eb6f9e1bc312"},"schema_version":"1.0","source":{"id":"2508.19980","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.19980","created_at":"2026-07-05T12:00:21Z"},{"alias_kind":"arxiv_version","alias_value":"2508.19980v1","created_at":"2026-07-05T12:00:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.19980","created_at":"2026-07-05T12:00:21Z"},{"alias_kind":"pith_short_12","alias_value":"URQGX36RMICX","created_at":"2026-07-05T12:00:21Z"},{"alias_kind":"pith_short_16","alias_value":"URQGX36RMICXEZ7R","created_at":"2026-07-05T12:00:21Z"},{"alias_kind":"pith_short_8","alias_value":"URQGX36R","created_at":"2026-07-05T12:00:21Z"}],"graph_snapshots":[{"event_id":"sha256:9796eb521df61a4f32b37f1a409f30dbf08c831426c920dff9377bbb6d1e8632","target":"graph","created_at":"2026-07-05T12:00:21Z","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/2508.19980/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As language models are increasingly deployed as autonomous agents in high-stakes settings, ensuring that they reliably follow user-defined rules has become a critical safety concern. To this end, we study whether language models exhibit contextual robustness, or the capability to adhere to context-dependent safety specifications. For this analysis, we develop a benchmark (PasswordEval) that measures whether language models can correctly determine when a user request is authorized (i.e., with a correct password). We find that current open- and closed-source models struggle with this seemingly s","authors_text":"Alexander Robey, Andy Zou, Dylan Sam, J. Zico Kolter, Matt Fredrikson","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-27T15:39:46Z","title":"Evaluating Language Model Reasoning about Confidential Information"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.19980","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:e0c7036e67f7dd44ef88250ebff099ac0d5a8fe1def118ed60ca3cd17fa35310","target":"record","created_at":"2026-07-05T12:00:21Z","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":"e25c8c3d37918987ec835d8cad138fa9629a1e66f593cdfe2ae583f52a9f8aff","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-27T15:39:46Z","title_canon_sha256":"48cdc9f8a2db77d92ba3e746711c7231cdb0304bcc2a2201c8b3eb6f9e1bc312"},"schema_version":"1.0","source":{"id":"2508.19980","kind":"arxiv","version":1}},"canonical_sha256":"a4606befd162057267f19b1134d73a754fcc3320ebf5e0bd5a2bde5920ecec21","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a4606befd162057267f19b1134d73a754fcc3320ebf5e0bd5a2bde5920ecec21","first_computed_at":"2026-07-05T12:00:21.175827Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:00:21.175827Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TvpgmLUl0JNVKw5pU53x23dMX68CiVzQ6BR90/FYnF31N/0ni2eKPOzJFSMl+fj/p4iKj5ZNDLLsGbxlpvTfDw==","signature_status":"signed_v1","signed_at":"2026-07-05T12:00:21.176215Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.19980","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e0c7036e67f7dd44ef88250ebff099ac0d5a8fe1def118ed60ca3cd17fa35310","sha256:9796eb521df61a4f32b37f1a409f30dbf08c831426c920dff9377bbb6d1e8632"],"state_sha256":"42d19bb3d04f3861375f88de571818de819c18152b6fa1e04d9fce0a59ab7f18"}