{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:W7RBYHSD3R3M4XZPMKAD37DQTR","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c3df21c8b1fc65304f69d40f0eadde7a6f2fdd1057862426824606d507acd0d6","cross_cats_sorted":["cs.CY","cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-04T12:30:13Z","title_canon_sha256":"22122b368065455334e4a6366445d64bc856289a9c01adfdf138552f125a653a"},"schema_version":"1.0","source":{"id":"2608.03569","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.03569","created_at":"2026-08-05T01:36:39Z"},{"alias_kind":"arxiv_version","alias_value":"2608.03569v1","created_at":"2026-08-05T01:36:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.03569","created_at":"2026-08-05T01:36:39Z"},{"alias_kind":"pith_short_12","alias_value":"W7RBYHSD3R3M","created_at":"2026-08-05T01:36:39Z"},{"alias_kind":"pith_short_16","alias_value":"W7RBYHSD3R3M4XZP","created_at":"2026-08-05T01:36:39Z"},{"alias_kind":"pith_short_8","alias_value":"W7RBYHSD","created_at":"2026-08-05T01:36:39Z"}],"graph_snapshots":[{"event_id":"sha256:803fd2ec223b215d12307d1c8261c8d34188512222976f6942e28139edbcf802","target":"graph","created_at":"2026-08-05T01:36:39Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2608.03569/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Benchmarking the ability of AI scientists to generate novel ideas is notoriously difficult. Existing benchmarks in this field have made progress in evaluating scientific reasoning and research replication, but often rely on synthetic tasks or retrospective targets, which may be confounded by prior exposure. We hypothesize that complex, adversarial, fast-moving real-world domains where expert practitioners independently generate observable outputs can provide a practical solution to fill this gap and evaluate the capabilities needed for AI scientists, including reasoning, novelty, and hypothesi","authors_text":"Philip Torr, William Bolton","cross_cats":["cs.CY","cs.LG","cs.MA"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-04T12:30:13Z","title":"Adversarial Fast-Moving Real-World Domains as Test Beds for Benchmarking AI Scientist Capabilities"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.03569","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:5373beb014d15ffad192f9167763ebd31140a10583779b91d92050920de2307b","target":"record","created_at":"2026-08-05T01:36:39Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c3df21c8b1fc65304f69d40f0eadde7a6f2fdd1057862426824606d507acd0d6","cross_cats_sorted":["cs.CY","cs.LG","cs.MA"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-04T12:30:13Z","title_canon_sha256":"22122b368065455334e4a6366445d64bc856289a9c01adfdf138552f125a653a"},"schema_version":"1.0","source":{"id":"2608.03569","kind":"arxiv","version":1}},"canonical_sha256":"b7e21c1e43dc76ce5f2f62803dfc709c62630c623f38764bd05f8eb12256705e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b7e21c1e43dc76ce5f2f62803dfc709c62630c623f38764bd05f8eb12256705e","first_computed_at":"2026-08-05T01:36:39.956600Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-05T01:36:39.956600Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"6aZ/Jayp2Yc4IItfdHQKC9FNiJgsWs/oxy2jbhXnK+zOOTLYzCZq2DDv0JUj8eTnxheNxK/eEr8SmxwufZD2Aw==","signature_status":"signed_v1","signed_at":"2026-08-05T01:36:39.958057Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.03569","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5373beb014d15ffad192f9167763ebd31140a10583779b91d92050920de2307b","sha256:803fd2ec223b215d12307d1c8261c8d34188512222976f6942e28139edbcf802"],"state_sha256":"34ebc1857880d7899426ccde63d27719641c79f446418a4f209f0ebe41decb1d"}