{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:QM3XCFOSCPCUFWE66RHNVZGJDS","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":"2fa74c08a3422f6dfda9950997c59ea8b1facf83935af23ec523580085523da3","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-26T18:25:26Z","title_canon_sha256":"40cf86c06d774b09f36b632e7f6601f47bc9f9f6f2010fe5b9ad932c0ebae663"},"schema_version":"1.0","source":{"id":"2501.15627","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.15627","created_at":"2026-07-05T10:59:05Z"},{"alias_kind":"arxiv_version","alias_value":"2501.15627v2","created_at":"2026-07-05T10:59:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.15627","created_at":"2026-07-05T10:59:05Z"},{"alias_kind":"pith_short_12","alias_value":"QM3XCFOSCPCU","created_at":"2026-07-05T10:59:05Z"},{"alias_kind":"pith_short_16","alias_value":"QM3XCFOSCPCUFWE6","created_at":"2026-07-05T10:59:05Z"},{"alias_kind":"pith_short_8","alias_value":"QM3XCFOS","created_at":"2026-07-05T10:59:05Z"}],"graph_snapshots":[{"event_id":"sha256:e6388d1bb0d2d59eb78e9dcc2c1c4ddde9a0aa790ce4f2b24042b4ce7eb13306","target":"graph","created_at":"2026-07-05T10:59:05Z","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/2501.15627/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present HardML, a benchmark designed to evaluate the knowledge and reasoning abilities in the fields of data science and machine learning. HardML comprises a diverse set of 100 challenging multiple-choice questions, handcrafted over a period of 6 months, covering the most popular and modern branches of data science and machine learning. These questions are challenging even for a typical Senior Machine Learning Engineer to answer correctly. To minimize the risk of data contamination, HardML uses mostly original content devised by the author. Current state of the art AI models achieve a 30% e","authors_text":"Tidor-Vlad Pricope","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-26T18:25:26Z","title":"HardML: A Benchmark For Evaluating Data Science And Machine Learning knowledge and reasoning in AI"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.15627","kind":"arxiv","version":2},"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:f3ad950e1463c8240c728c52b28a9aca2677ebb6201ea9266815fd41bf9b00d8","target":"record","created_at":"2026-07-05T10:59:05Z","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":"2fa74c08a3422f6dfda9950997c59ea8b1facf83935af23ec523580085523da3","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-26T18:25:26Z","title_canon_sha256":"40cf86c06d774b09f36b632e7f6601f47bc9f9f6f2010fe5b9ad932c0ebae663"},"schema_version":"1.0","source":{"id":"2501.15627","kind":"arxiv","version":2}},"canonical_sha256":"83377115d213c542d89ef44edae4c91cb5e7e4688c354b1a087866b35d0c9a2f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"83377115d213c542d89ef44edae4c91cb5e7e4688c354b1a087866b35d0c9a2f","first_computed_at":"2026-07-05T10:59:05.145434Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:59:05.145434Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JGG8xxaZFRSg6nljuGoLSRzpAWzdIaPM/H4p9L8iDv2HKOFMN95mkmGOWvUw0WNzPBJwvZVXj3pNyfIHUfRFAw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:59:05.145906Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.15627","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f3ad950e1463c8240c728c52b28a9aca2677ebb6201ea9266815fd41bf9b00d8","sha256:e6388d1bb0d2d59eb78e9dcc2c1c4ddde9a0aa790ce4f2b24042b4ce7eb13306"],"state_sha256":"9a31a9f8da526bb2ae9a848d65c8aabd6ad454e82e8941571c40c93898dba0f2"}