{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:R2OO6OPTXKV23UZE7SJDPQ54M2","short_pith_number":"pith:R2OO6OPT","schema_version":"1.0","canonical_sha256":"8e9cef39f3baabadd324fc9237c3bc66b3e35e3a4b41660eb1419955a17998cc","source":{"kind":"arxiv","id":"2508.05441","version":1},"attestation_state":"computed","paper":{"title":"Tail-Risk-Safe Monte Carlo Tree Search under PAC-Level Guarantees","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Arnob Ghosh, Tian Lan, Zuyuan Zhang","submitted_at":"2025-08-07T14:31:22Z","abstract_excerpt":"Making decisions with respect to just the expected returns in Monte Carlo Tree Search (MCTS) cannot account for the potential range of high-risk, adverse outcomes associated with a decision. To this end, safety-aware MCTS often consider some constrained variants -- by introducing some form of mean risk measures or hard cost thresholds. These approaches fail to provide rigorous tail-safety guarantees with respect to extreme or high-risk outcomes (denoted as tail-risk), potentially resulting in serious consequence in high-stake scenarios. This paper addresses the problem by developing two novel "},"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":"2508.05441","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-07T14:31:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0081c10fac01a81aca9ed2250cc52f099dfa8ec915feaec9b17c49e435018f26","abstract_canon_sha256":"0c74b762ea2a4e93656a6f5bb6285d71c696dce2da958178bbd1105b2f0f687d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:50:13.330886Z","signature_b64":"ZDUe0c9hRwAJGeuYNFPe0UgyI+zpTXM0kxnHmwBB5xl/3xNbmfY+4OtXku9VKHumHXBIuSoFkGwvZd+Pj0+jBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8e9cef39f3baabadd324fc9237c3bc66b3e35e3a4b41660eb1419955a17998cc","last_reissued_at":"2026-07-05T11:50:13.330413Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:50:13.330413Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tail-Risk-Safe Monte Carlo Tree Search under PAC-Level Guarantees","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Arnob Ghosh, Tian Lan, Zuyuan Zhang","submitted_at":"2025-08-07T14:31:22Z","abstract_excerpt":"Making decisions with respect to just the expected returns in Monte Carlo Tree Search (MCTS) cannot account for the potential range of high-risk, adverse outcomes associated with a decision. To this end, safety-aware MCTS often consider some constrained variants -- by introducing some form of mean risk measures or hard cost thresholds. These approaches fail to provide rigorous tail-safety guarantees with respect to extreme or high-risk outcomes (denoted as tail-risk), potentially resulting in serious consequence in high-stake scenarios. This paper addresses the problem by developing two novel "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.05441","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/2508.05441/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":"2508.05441","created_at":"2026-07-05T11:50:13.330474+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.05441v1","created_at":"2026-07-05T11:50:13.330474+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.05441","created_at":"2026-07-05T11:50:13.330474+00:00"},{"alias_kind":"pith_short_12","alias_value":"R2OO6OPTXKV2","created_at":"2026-07-05T11:50:13.330474+00:00"},{"alias_kind":"pith_short_16","alias_value":"R2OO6OPTXKV23UZE","created_at":"2026-07-05T11:50:13.330474+00:00"},{"alias_kind":"pith_short_8","alias_value":"R2OO6OPT","created_at":"2026-07-05T11:50:13.330474+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.06500","citing_title":"Operator-Guided Invariance Learning for Continuous Reinforcement Learning","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R2OO6OPTXKV23UZE7SJDPQ54M2","json":"https://pith.science/pith/R2OO6OPTXKV23UZE7SJDPQ54M2.json","graph_json":"https://pith.science/api/pith-number/R2OO6OPTXKV23UZE7SJDPQ54M2/graph.json","events_json":"https://pith.science/api/pith-number/R2OO6OPTXKV23UZE7SJDPQ54M2/events.json","paper":"https://pith.science/paper/R2OO6OPT"},"agent_actions":{"view_html":"https://pith.science/pith/R2OO6OPTXKV23UZE7SJDPQ54M2","download_json":"https://pith.science/pith/R2OO6OPTXKV23UZE7SJDPQ54M2.json","view_paper":"https://pith.science/paper/R2OO6OPT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.05441&json=true","fetch_graph":"https://pith.science/api/pith-number/R2OO6OPTXKV23UZE7SJDPQ54M2/graph.json","fetch_events":"https://pith.science/api/pith-number/R2OO6OPTXKV23UZE7SJDPQ54M2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R2OO6OPTXKV23UZE7SJDPQ54M2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R2OO6OPTXKV23UZE7SJDPQ54M2/action/storage_attestation","attest_author":"https://pith.science/pith/R2OO6OPTXKV23UZE7SJDPQ54M2/action/author_attestation","sign_citation":"https://pith.science/pith/R2OO6OPTXKV23UZE7SJDPQ54M2/action/citation_signature","submit_replication":"https://pith.science/pith/R2OO6OPTXKV23UZE7SJDPQ54M2/action/replication_record"}},"created_at":"2026-07-05T11:50:13.330474+00:00","updated_at":"2026-07-05T11:50:13.330474+00:00"}