{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:45BMV5WIHHLC5AXUUSS4UZLHOB","short_pith_number":"pith:45BMV5WI","schema_version":"1.0","canonical_sha256":"e742caf6c839d62e82f4a4a5ca6567707b6b13862b1b861ecf48919470685b6e","source":{"kind":"arxiv","id":"2606.28374","version":1},"attestation_state":"computed","paper":{"title":"Recursive Self-Evolving Agents via Held-Out Selection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Michael Nguyen, Paul Vuong, Quoc Nguyen","submitted_at":"2026-06-17T14:53:36Z","abstract_excerpt":"LLM agents are increasingly improved without weight updates by evolving a natural-language artifact, such as reflections, workflows, playbooks, cheatsheets, or optimized prompts, that conditions a frozen policy. Such methods are typically reported as wins on the single benchmark where they help. We study them apples-to-apples and surface a sharper picture. We introduce RSEA, a Recursive Self-Evolving Agent that carries a compact three-layer natural-language state: an imperative strategy, reusable skills, and a procedural playbook. Across generations, RSEA rewrites all three layers from its own"},"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":"2606.28374","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-17T14:53:36Z","cross_cats_sorted":[],"title_canon_sha256":"364075c6f789deaabe5b15e6a3414d4f24150061c810ca7af63443f03bcf95ce","abstract_canon_sha256":"f18e83d3768a4eb4c3b293a8746a643adda72d6a1c17f0ee607994a3821b9fbe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T00:15:12.322721Z","signature_b64":"hrPQWbptku3hhh1CTEuF8wVR9KBhxOQpT4hN/VQQHSi+76PejWbh0dbCIQLH6oRgSrOpNEGAp6l0bKGRAwz+BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e742caf6c839d62e82f4a4a5ca6567707b6b13862b1b861ecf48919470685b6e","last_reissued_at":"2026-06-30T00:15:12.322231Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T00:15:12.322231Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Recursive Self-Evolving Agents via Held-Out Selection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Michael Nguyen, Paul Vuong, Quoc Nguyen","submitted_at":"2026-06-17T14:53:36Z","abstract_excerpt":"LLM agents are increasingly improved without weight updates by evolving a natural-language artifact, such as reflections, workflows, playbooks, cheatsheets, or optimized prompts, that conditions a frozen policy. Such methods are typically reported as wins on the single benchmark where they help. We study them apples-to-apples and surface a sharper picture. We introduce RSEA, a Recursive Self-Evolving Agent that carries a compact three-layer natural-language state: an imperative strategy, reusable skills, and a procedural playbook. Across generations, RSEA rewrites all three layers from its own"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.28374","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/2606.28374/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":"2606.28374","created_at":"2026-06-30T00:15:12.322289+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.28374v1","created_at":"2026-06-30T00:15:12.322289+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.28374","created_at":"2026-06-30T00:15:12.322289+00:00"},{"alias_kind":"pith_short_12","alias_value":"45BMV5WIHHLC","created_at":"2026-06-30T00:15:12.322289+00:00"},{"alias_kind":"pith_short_16","alias_value":"45BMV5WIHHLC5AXU","created_at":"2026-06-30T00:15:12.322289+00:00"},{"alias_kind":"pith_short_8","alias_value":"45BMV5WI","created_at":"2026-06-30T00:15:12.322289+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/45BMV5WIHHLC5AXUUSS4UZLHOB","json":"https://pith.science/pith/45BMV5WIHHLC5AXUUSS4UZLHOB.json","graph_json":"https://pith.science/api/pith-number/45BMV5WIHHLC5AXUUSS4UZLHOB/graph.json","events_json":"https://pith.science/api/pith-number/45BMV5WIHHLC5AXUUSS4UZLHOB/events.json","paper":"https://pith.science/paper/45BMV5WI"},"agent_actions":{"view_html":"https://pith.science/pith/45BMV5WIHHLC5AXUUSS4UZLHOB","download_json":"https://pith.science/pith/45BMV5WIHHLC5AXUUSS4UZLHOB.json","view_paper":"https://pith.science/paper/45BMV5WI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.28374&json=true","fetch_graph":"https://pith.science/api/pith-number/45BMV5WIHHLC5AXUUSS4UZLHOB/graph.json","fetch_events":"https://pith.science/api/pith-number/45BMV5WIHHLC5AXUUSS4UZLHOB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/45BMV5WIHHLC5AXUUSS4UZLHOB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/45BMV5WIHHLC5AXUUSS4UZLHOB/action/storage_attestation","attest_author":"https://pith.science/pith/45BMV5WIHHLC5AXUUSS4UZLHOB/action/author_attestation","sign_citation":"https://pith.science/pith/45BMV5WIHHLC5AXUUSS4UZLHOB/action/citation_signature","submit_replication":"https://pith.science/pith/45BMV5WIHHLC5AXUUSS4UZLHOB/action/replication_record"}},"created_at":"2026-06-30T00:15:12.322289+00:00","updated_at":"2026-06-30T00:15:12.322289+00:00"}