{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DJXFP4GSUQSBDRU4ZKHO4KCQSF","short_pith_number":"pith:DJXFP4GS","schema_version":"1.0","canonical_sha256":"1a6e57f0d2a42411c69cca8eee2850916c7b6bde4242655ac15ca3fa2ce58be9","source":{"kind":"arxiv","id":"2606.08106","version":1},"attestation_state":"computed","paper":{"title":"PACE: Anytime-Valid Acceptance Tests for Self-Evolving Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.AI","authors_text":"Zayx Shawn","submitted_at":"2026-06-06T11:12:11Z","abstract_excerpt":"Self-evolving agents improve by repeatedly proposing changes to their own prompts, skills, or workflows and keeping those that score higher on a small held-out set. Almost all effort has gone into the proposer that generates candidates; we argue the weak point is the acceptor, the rule that decides whether to commit a change. Applied hundreds of times against the same noisy dev estimate, the ubiquitous \"keep it if the score went up\" rule is uncontrolled adaptive multiple testing: the agent effectively p-hacks itself, accumulating false commits that make it churn and drift rather than improve.\n"},"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":"2606.08106","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-06T11:12:11Z","cross_cats_sorted":["cs.MA"],"title_canon_sha256":"358dc4612b226ecfc46438d6007491e081c4c31a0c8d202bda06d534e5bb9d4c","abstract_canon_sha256":"9545b07a15239903275139375ae6f84351734ab5ac560150a74b85886dd345df"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:26.761842Z","signature_b64":"pX4XSQnoskTnn5pwbo2Gh4rNbbJ8ti899ddxr5MHPPALA+82SF2gjcAVANGudXyQ6GzNtkrKw24VwXDbSMLVAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1a6e57f0d2a42411c69cca8eee2850916c7b6bde4242655ac15ca3fa2ce58be9","last_reissued_at":"2026-06-09T01:05:26.761230Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:26.761230Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PACE: Anytime-Valid Acceptance Tests for Self-Evolving Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.MA"],"primary_cat":"cs.AI","authors_text":"Zayx Shawn","submitted_at":"2026-06-06T11:12:11Z","abstract_excerpt":"Self-evolving agents improve by repeatedly proposing changes to their own prompts, skills, or workflows and keeping those that score higher on a small held-out set. Almost all effort has gone into the proposer that generates candidates; we argue the weak point is the acceptor, the rule that decides whether to commit a change. Applied hundreds of times against the same noisy dev estimate, the ubiquitous \"keep it if the score went up\" rule is uncontrolled adaptive multiple testing: the agent effectively p-hacks itself, accumulating false commits that make it churn and drift rather than improve.\n"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.08106","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/2606.08106/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":"2606.08106","created_at":"2026-06-09T01:05:26.761325+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.08106v1","created_at":"2026-06-09T01:05:26.761325+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.08106","created_at":"2026-06-09T01:05:26.761325+00:00"},{"alias_kind":"pith_short_12","alias_value":"DJXFP4GSUQSB","created_at":"2026-06-09T01:05:26.761325+00:00"},{"alias_kind":"pith_short_16","alias_value":"DJXFP4GSUQSBDRU4","created_at":"2026-06-09T01:05:26.761325+00:00"},{"alias_kind":"pith_short_8","alias_value":"DJXFP4GS","created_at":"2026-06-09T01:05:26.761325+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/DJXFP4GSUQSBDRU4ZKHO4KCQSF","json":"https://pith.science/pith/DJXFP4GSUQSBDRU4ZKHO4KCQSF.json","graph_json":"https://pith.science/api/pith-number/DJXFP4GSUQSBDRU4ZKHO4KCQSF/graph.json","events_json":"https://pith.science/api/pith-number/DJXFP4GSUQSBDRU4ZKHO4KCQSF/events.json","paper":"https://pith.science/paper/DJXFP4GS"},"agent_actions":{"view_html":"https://pith.science/pith/DJXFP4GSUQSBDRU4ZKHO4KCQSF","download_json":"https://pith.science/pith/DJXFP4GSUQSBDRU4ZKHO4KCQSF.json","view_paper":"https://pith.science/paper/DJXFP4GS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.08106&json=true","fetch_graph":"https://pith.science/api/pith-number/DJXFP4GSUQSBDRU4ZKHO4KCQSF/graph.json","fetch_events":"https://pith.science/api/pith-number/DJXFP4GSUQSBDRU4ZKHO4KCQSF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DJXFP4GSUQSBDRU4ZKHO4KCQSF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DJXFP4GSUQSBDRU4ZKHO4KCQSF/action/storage_attestation","attest_author":"https://pith.science/pith/DJXFP4GSUQSBDRU4ZKHO4KCQSF/action/author_attestation","sign_citation":"https://pith.science/pith/DJXFP4GSUQSBDRU4ZKHO4KCQSF/action/citation_signature","submit_replication":"https://pith.science/pith/DJXFP4GSUQSBDRU4ZKHO4KCQSF/action/replication_record"}},"created_at":"2026-06-09T01:05:26.761325+00:00","updated_at":"2026-06-09T01:05:26.761325+00:00"}