{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:UEFLSXGTK75Z4BG4SOSNNQY5XA","short_pith_number":"pith:UEFLSXGT","schema_version":"1.0","canonical_sha256":"a10ab95cd357fb9e04dc93a4d6c31db834d1ea31acd81d5c56694f17ae311598","source":{"kind":"arxiv","id":"2605.18580","version":1},"attestation_state":"computed","paper":{"title":"When Outcome Looks Right But Discipline Fails: Trace-Based Evaluation Under Hidden Competitor State","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Peiying Zhu, Sidi Chang","submitted_at":"2026-05-18T15:58:34Z","abstract_excerpt":"Outcome-only evaluation can certify economically unsafe agents: a policy can hit a business KPI while violating deployable behavioral discipline. In hotel pricing with hidden competitor state, a learner can achieve plausible revenue per available room while failing to preserve the rate discipline of a rule-based revenue-management competitor.\n  We introduce discipline stability, a trace-based evaluation paradigm: define the benchmark behavior, restrict observations to the deployment regime, induce trace diagnostics from failure, separate mechanisms with ablations, and test transfer and deploym"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.18580","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-18T15:58:34Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6e77b0d23055651aeb42ce0d3690fa08a87f9df44650ee1c52930038e242d72d","abstract_canon_sha256":"340811ac4f75d8c6aa663a6a2a7371307ed6389e9edfd814a312477c4c5ce94a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:06:08.798606Z","signature_b64":"DNEPxIr8d3aAGcLohOmDTkKUxuPJhfyhhoMFgsWss3a4g7ZMww3Zw5EoIkScN3jFspjBGk3L+WOYdhqYQ272Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a10ab95cd357fb9e04dc93a4d6c31db834d1ea31acd81d5c56694f17ae311598","last_reissued_at":"2026-05-20T00:06:08.797299Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:06:08.797299Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When Outcome Looks Right But Discipline Fails: Trace-Based Evaluation Under Hidden Competitor State","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Peiying Zhu, Sidi Chang","submitted_at":"2026-05-18T15:58:34Z","abstract_excerpt":"Outcome-only evaluation can certify economically unsafe agents: a policy can hit a business KPI while violating deployable behavioral discipline. In hotel pricing with hidden competitor state, a learner can achieve plausible revenue per available room while failing to preserve the rate discipline of a rule-based revenue-management competitor.\n  We introduce discipline stability, a trace-based evaluation paradigm: define the benchmark behavior, restrict observations to the deployment regime, induce trace diagnostics from failure, separate mechanisms with ablations, and test transfer and deploym"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18580","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.18580/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T00:01:59.313110Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"9c9283c18b4b270adf12b20df92a87c893d6240edbf34e1f4f46a3c6fcf644fc"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.18580","created_at":"2026-05-20T00:06:08.797480+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.18580v1","created_at":"2026-05-20T00:06:08.797480+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18580","created_at":"2026-05-20T00:06:08.797480+00:00"},{"alias_kind":"pith_short_12","alias_value":"UEFLSXGTK75Z","created_at":"2026-05-20T00:06:08.797480+00:00"},{"alias_kind":"pith_short_16","alias_value":"UEFLSXGTK75Z4BG4","created_at":"2026-05-20T00:06:08.797480+00:00"},{"alias_kind":"pith_short_8","alias_value":"UEFLSXGT","created_at":"2026-05-20T00:06:08.797480+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UEFLSXGTK75Z4BG4SOSNNQY5XA","json":"https://pith.science/pith/UEFLSXGTK75Z4BG4SOSNNQY5XA.json","graph_json":"https://pith.science/api/pith-number/UEFLSXGTK75Z4BG4SOSNNQY5XA/graph.json","events_json":"https://pith.science/api/pith-number/UEFLSXGTK75Z4BG4SOSNNQY5XA/events.json","paper":"https://pith.science/paper/UEFLSXGT"},"agent_actions":{"view_html":"https://pith.science/pith/UEFLSXGTK75Z4BG4SOSNNQY5XA","download_json":"https://pith.science/pith/UEFLSXGTK75Z4BG4SOSNNQY5XA.json","view_paper":"https://pith.science/paper/UEFLSXGT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.18580&json=true","fetch_graph":"https://pith.science/api/pith-number/UEFLSXGTK75Z4BG4SOSNNQY5XA/graph.json","fetch_events":"https://pith.science/api/pith-number/UEFLSXGTK75Z4BG4SOSNNQY5XA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UEFLSXGTK75Z4BG4SOSNNQY5XA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UEFLSXGTK75Z4BG4SOSNNQY5XA/action/storage_attestation","attest_author":"https://pith.science/pith/UEFLSXGTK75Z4BG4SOSNNQY5XA/action/author_attestation","sign_citation":"https://pith.science/pith/UEFLSXGTK75Z4BG4SOSNNQY5XA/action/citation_signature","submit_replication":"https://pith.science/pith/UEFLSXGTK75Z4BG4SOSNNQY5XA/action/replication_record"}},"created_at":"2026-05-20T00:06:08.797480+00:00","updated_at":"2026-05-20T00:06:08.797480+00:00"}