{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SBCF5OZ5DWBCIYUSZCLSPNWEMK","short_pith_number":"pith:SBCF5OZ5","schema_version":"1.0","canonical_sha256":"90445ebb3d1d82246292c89727b6c46289436fa0600be68b93ab7a60f35a61a4","source":{"kind":"arxiv","id":"2608.06362","version":1},"attestation_state":"computed","paper":{"title":"AV-AIVAT: 74x Cheaper Agent Evaluation with Certified Anytime-Valid Stopping in Imperfect-Information Games","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.MA"],"primary_cat":"cs.GT","authors_text":"Boning Li, Longbo Huang, Yu Chen","submitted_at":"2026-08-06T17:57:11Z","abstract_excerpt":"Deciding which of two agents is stronger means playing games until skill outweighs luck, and every game costs money, model inference, or expert time. Since the number of games needed is unknown, fixed-budget evaluations either keep paying after the result is settled or stop before the agents can be told apart, while naive optional stopping with an ordinary confidence interval invalidates the stated level. We make such an evaluation stop as soon as its evidence suffices, with the guarantee intact. The Action-Informed Value Assessment Tool (AIVAT) reduces variance in imperfect-information games "},"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":"2608.06362","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.GT","submitted_at":"2026-08-06T17:57:11Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG","cs.MA"],"title_canon_sha256":"c8cb330351388302e6241861673ec4507ceb3f87563742c1fab742f0f7e32a56","abstract_canon_sha256":"a593fb240ff89c561d56aa9eb5e1b7a842201fc4c597c43da8e5c9ffe67be522"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-07T01:41:29.025001Z","signature_b64":"TL6L08vjnTOHxtKC94kx6c4zUXiRvcdfxG9wki/RP2pSgrXes4t6bj1d36qDpkPvt+4TDjohISw+3QaD0xPiDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"90445ebb3d1d82246292c89727b6c46289436fa0600be68b93ab7a60f35a61a4","last_reissued_at":"2026-08-07T01:41:29.023386Z","signature_status":"signed_v1","first_computed_at":"2026-08-07T01:41:29.023386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AV-AIVAT: 74x Cheaper Agent Evaluation with Certified Anytime-Valid Stopping in Imperfect-Information Games","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.MA"],"primary_cat":"cs.GT","authors_text":"Boning Li, Longbo Huang, Yu Chen","submitted_at":"2026-08-06T17:57:11Z","abstract_excerpt":"Deciding which of two agents is stronger means playing games until skill outweighs luck, and every game costs money, model inference, or expert time. Since the number of games needed is unknown, fixed-budget evaluations either keep paying after the result is settled or stop before the agents can be told apart, while naive optional stopping with an ordinary confidence interval invalidates the stated level. We make such an evaluation stop as soon as its evidence suffices, with the guarantee intact. The Action-Informed Value Assessment Tool (AIVAT) reduces variance in imperfect-information games "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.06362","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/2608.06362/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":"2608.06362","created_at":"2026-08-07T01:41:29.025963+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.06362v1","created_at":"2026-08-07T01:41:29.025963+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.06362","created_at":"2026-08-07T01:41:29.025963+00:00"},{"alias_kind":"pith_short_12","alias_value":"SBCF5OZ5DWBC","created_at":"2026-08-07T01:41:29.025963+00:00"},{"alias_kind":"pith_short_16","alias_value":"SBCF5OZ5DWBCIYUS","created_at":"2026-08-07T01:41:29.025963+00:00"},{"alias_kind":"pith_short_8","alias_value":"SBCF5OZ5","created_at":"2026-08-07T01:41:29.025963+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/SBCF5OZ5DWBCIYUSZCLSPNWEMK","json":"https://pith.science/pith/SBCF5OZ5DWBCIYUSZCLSPNWEMK.json","graph_json":"https://pith.science/api/pith-number/SBCF5OZ5DWBCIYUSZCLSPNWEMK/graph.json","events_json":"https://pith.science/api/pith-number/SBCF5OZ5DWBCIYUSZCLSPNWEMK/events.json","paper":"https://pith.science/paper/SBCF5OZ5"},"agent_actions":{"view_html":"https://pith.science/pith/SBCF5OZ5DWBCIYUSZCLSPNWEMK","download_json":"https://pith.science/pith/SBCF5OZ5DWBCIYUSZCLSPNWEMK.json","view_paper":"https://pith.science/paper/SBCF5OZ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.06362&json=true","fetch_graph":"https://pith.science/api/pith-number/SBCF5OZ5DWBCIYUSZCLSPNWEMK/graph.json","fetch_events":"https://pith.science/api/pith-number/SBCF5OZ5DWBCIYUSZCLSPNWEMK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SBCF5OZ5DWBCIYUSZCLSPNWEMK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SBCF5OZ5DWBCIYUSZCLSPNWEMK/action/storage_attestation","attest_author":"https://pith.science/pith/SBCF5OZ5DWBCIYUSZCLSPNWEMK/action/author_attestation","sign_citation":"https://pith.science/pith/SBCF5OZ5DWBCIYUSZCLSPNWEMK/action/citation_signature","submit_replication":"https://pith.science/pith/SBCF5OZ5DWBCIYUSZCLSPNWEMK/action/replication_record"}},"created_at":"2026-08-07T01:41:29.025963+00:00","updated_at":"2026-08-07T01:41:29.025963+00:00"}