{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OV5DAM6DZXYBC2TPZRZJL3KEI6","short_pith_number":"pith:OV5DAM6D","schema_version":"1.0","canonical_sha256":"757a3033c3cdf0116a6fcc7295ed4447ab6741b23a0ef91f048e3a813068ca6d","source":{"kind":"arxiv","id":"2607.24300","version":1},"attestation_state":"computed","paper":{"title":"Self-Authored Verification Is Unreliable in Heuristic Self-Improving Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.CL","authors_text":"Cong Cao, Dakui Wang, Diandian Guo, Fangfang Yuan, Yingqi Wang, Yueshan Wang","submitted_at":"2026-07-27T11:45:14Z","abstract_excerpt":"Self-improving agents accumulate capability by repeatedly rewriting procedural policies, controllers, or heuristic rules. They typically rely on self-authored tests or metrics to decide whether to accept subsequent edits. The agent controls both the optimized object and its verifier. As a result, self-assigned scores can remain near perfect while real deployment performance degrades or stays low. We study this problem through the verifier--deployment gap. This gap refers to the discrepancy between an agent's self-authored verification signal and a sealed deployment evaluation that the agent ca"},"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":"2607.24300","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-27T11:45:14Z","cross_cats_sorted":["cs.MA"],"title_canon_sha256":"03b39d012488317f4dcc1fdaa74c521d23646292af24f8271b353b6b87bbb08c","abstract_canon_sha256":"b23c736c8f3a2dd4c5dda99f6a2c7d94c5b2ebf4d1c907242d83ab98d83cf1fe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T02:23:56.232205Z","signature_b64":"TuIfs1hDNS/OfmsPtLI5eUlOuF+o5qNovqNMV4aLwI+P+ZoUE4rjemPs/EkiOrhh+JDuHNiAGgJKjO9ygn+rAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"757a3033c3cdf0116a6fcc7295ed4447ab6741b23a0ef91f048e3a813068ca6d","last_reissued_at":"2026-07-28T02:23:56.231177Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T02:23:56.231177Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Authored Verification Is Unreliable in Heuristic Self-Improving Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.CL","authors_text":"Cong Cao, Dakui Wang, Diandian Guo, Fangfang Yuan, Yingqi Wang, Yueshan Wang","submitted_at":"2026-07-27T11:45:14Z","abstract_excerpt":"Self-improving agents accumulate capability by repeatedly rewriting procedural policies, controllers, or heuristic rules. They typically rely on self-authored tests or metrics to decide whether to accept subsequent edits. The agent controls both the optimized object and its verifier. As a result, self-assigned scores can remain near perfect while real deployment performance degrades or stays low. We study this problem through the verifier--deployment gap. This gap refers to the discrepancy between an agent's self-authored verification signal and a sealed deployment evaluation that the agent ca"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.24300","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/2607.24300/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":"2607.24300","created_at":"2026-07-28T02:23:56.231664+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.24300v1","created_at":"2026-07-28T02:23:56.231664+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.24300","created_at":"2026-07-28T02:23:56.231664+00:00"},{"alias_kind":"pith_short_12","alias_value":"OV5DAM6DZXYB","created_at":"2026-07-28T02:23:56.231664+00:00"},{"alias_kind":"pith_short_16","alias_value":"OV5DAM6DZXYBC2TP","created_at":"2026-07-28T02:23:56.231664+00:00"},{"alias_kind":"pith_short_8","alias_value":"OV5DAM6D","created_at":"2026-07-28T02:23:56.231664+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/OV5DAM6DZXYBC2TPZRZJL3KEI6","json":"https://pith.science/pith/OV5DAM6DZXYBC2TPZRZJL3KEI6.json","graph_json":"https://pith.science/api/pith-number/OV5DAM6DZXYBC2TPZRZJL3KEI6/graph.json","events_json":"https://pith.science/api/pith-number/OV5DAM6DZXYBC2TPZRZJL3KEI6/events.json","paper":"https://pith.science/paper/OV5DAM6D"},"agent_actions":{"view_html":"https://pith.science/pith/OV5DAM6DZXYBC2TPZRZJL3KEI6","download_json":"https://pith.science/pith/OV5DAM6DZXYBC2TPZRZJL3KEI6.json","view_paper":"https://pith.science/paper/OV5DAM6D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.24300&json=true","fetch_graph":"https://pith.science/api/pith-number/OV5DAM6DZXYBC2TPZRZJL3KEI6/graph.json","fetch_events":"https://pith.science/api/pith-number/OV5DAM6DZXYBC2TPZRZJL3KEI6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OV5DAM6DZXYBC2TPZRZJL3KEI6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OV5DAM6DZXYBC2TPZRZJL3KEI6/action/storage_attestation","attest_author":"https://pith.science/pith/OV5DAM6DZXYBC2TPZRZJL3KEI6/action/author_attestation","sign_citation":"https://pith.science/pith/OV5DAM6DZXYBC2TPZRZJL3KEI6/action/citation_signature","submit_replication":"https://pith.science/pith/OV5DAM6DZXYBC2TPZRZJL3KEI6/action/replication_record"}},"created_at":"2026-07-28T02:23:56.231664+00:00","updated_at":"2026-07-28T02:23:56.231664+00:00"}