{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IKRKQWDIG2H7OHPH5FGOITMRZP","short_pith_number":"pith:IKRKQWDI","schema_version":"1.0","canonical_sha256":"42a2a85868368ff71de7e94ce44d91cbff50845d9078f84c3829bedb2eaff58e","source":{"kind":"arxiv","id":"2605.20270","version":1},"attestation_state":"computed","paper":{"title":"Conformal Selective Acting: Anytime-Valid Risk Control for RLVR-Trained LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Hamed Khosravi, Xiaoming Huo","submitted_at":"2026-05-18T22:20:32Z","abstract_excerpt":"A local specialist LLM, fine-tuned with reinforcement learning from verifiable rewards (RLVR) on operator-local data, is installed in a regulated organization with per-deployment error budget $\\alpha$. The operator needs a safety certificate for this deployment's stream at every round: no pooling across deployments, no waiting for a long-run average. Existing wrappers cannot deliver this on adaptive, online-updated streams: offline conformal-risk methods require exchangeability; online-conformal methods bound only long-run averages; non-exchangeable extensions are marginally valid; and the clo"},"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":"2605.20270","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-18T22:20:32Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"54c11cb15f5a0a299b6e794df7cfd230065879a05344c85e82aa5f9709478572","abstract_canon_sha256":"eeea51e3295f6fbe7d1e43b8430e92589644828546f75acf34920126e5d4d158"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T00:04:23.476428Z","signature_b64":"906jrNbgPUlYsAM+5ispt9IgNE/VOjYNOMhB8vC6eEXFBgqKRIFcOHpWg71VL/IlKoS6QjgOffbqmfEE+aBEBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"42a2a85868368ff71de7e94ce44d91cbff50845d9078f84c3829bedb2eaff58e","last_reissued_at":"2026-05-21T00:04:23.475948Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T00:04:23.475948Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Conformal Selective Acting: Anytime-Valid Risk Control for RLVR-Trained LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Hamed Khosravi, Xiaoming Huo","submitted_at":"2026-05-18T22:20:32Z","abstract_excerpt":"A local specialist LLM, fine-tuned with reinforcement learning from verifiable rewards (RLVR) on operator-local data, is installed in a regulated organization with per-deployment error budget $\\alpha$. The operator needs a safety certificate for this deployment's stream at every round: no pooling across deployments, no waiting for a long-run average. Existing wrappers cannot deliver this on adaptive, online-updated streams: offline conformal-risk methods require exchangeability; online-conformal methods bound only long-run averages; non-exchangeable extensions are marginally valid; and the clo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.20270","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/2605.20270/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":"2605.20270","created_at":"2026-05-21T00:04:23.476009+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.20270v1","created_at":"2026-05-21T00:04:23.476009+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.20270","created_at":"2026-05-21T00:04:23.476009+00:00"},{"alias_kind":"pith_short_12","alias_value":"IKRKQWDIG2H7","created_at":"2026-05-21T00:04:23.476009+00:00"},{"alias_kind":"pith_short_16","alias_value":"IKRKQWDIG2H7OHPH","created_at":"2026-05-21T00:04:23.476009+00:00"},{"alias_kind":"pith_short_8","alias_value":"IKRKQWDI","created_at":"2026-05-21T00:04:23.476009+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/IKRKQWDIG2H7OHPH5FGOITMRZP","json":"https://pith.science/pith/IKRKQWDIG2H7OHPH5FGOITMRZP.json","graph_json":"https://pith.science/api/pith-number/IKRKQWDIG2H7OHPH5FGOITMRZP/graph.json","events_json":"https://pith.science/api/pith-number/IKRKQWDIG2H7OHPH5FGOITMRZP/events.json","paper":"https://pith.science/paper/IKRKQWDI"},"agent_actions":{"view_html":"https://pith.science/pith/IKRKQWDIG2H7OHPH5FGOITMRZP","download_json":"https://pith.science/pith/IKRKQWDIG2H7OHPH5FGOITMRZP.json","view_paper":"https://pith.science/paper/IKRKQWDI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.20270&json=true","fetch_graph":"https://pith.science/api/pith-number/IKRKQWDIG2H7OHPH5FGOITMRZP/graph.json","fetch_events":"https://pith.science/api/pith-number/IKRKQWDIG2H7OHPH5FGOITMRZP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IKRKQWDIG2H7OHPH5FGOITMRZP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IKRKQWDIG2H7OHPH5FGOITMRZP/action/storage_attestation","attest_author":"https://pith.science/pith/IKRKQWDIG2H7OHPH5FGOITMRZP/action/author_attestation","sign_citation":"https://pith.science/pith/IKRKQWDIG2H7OHPH5FGOITMRZP/action/citation_signature","submit_replication":"https://pith.science/pith/IKRKQWDIG2H7OHPH5FGOITMRZP/action/replication_record"}},"created_at":"2026-05-21T00:04:23.476009+00:00","updated_at":"2026-05-21T00:04:23.476009+00:00"}