{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:6OYI7A3L5PFGPS72FGYFUUNV5R","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":"11218bff621513be954c37ffef7677968450907c7af7b341d4bd343e8fe10a2c","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-11T20:46:56Z","title_canon_sha256":"fde39da6ddb8ad6edd159b5f82ab961732364b12bad98f856610673f0518841d"},"schema_version":"1.0","source":{"id":"2410.09247","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.09247","created_at":"2026-07-05T09:20:05Z"},{"alias_kind":"arxiv_version","alias_value":"2410.09247v1","created_at":"2026-07-05T09:20:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.09247","created_at":"2026-07-05T09:20:05Z"},{"alias_kind":"pith_short_12","alias_value":"6OYI7A3L5PFG","created_at":"2026-07-05T09:20:05Z"},{"alias_kind":"pith_short_16","alias_value":"6OYI7A3L5PFGPS72","created_at":"2026-07-05T09:20:05Z"},{"alias_kind":"pith_short_8","alias_value":"6OYI7A3L","created_at":"2026-07-05T09:20:05Z"}],"graph_snapshots":[{"event_id":"sha256:422db9398ce26f4b29c45e9b05bd784b4e7b017a22455c207f8cb226f1741aa5","target":"graph","created_at":"2026-07-05T09:20:05Z","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/2410.09247/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The training data for many Large Language Models (LLMs) is contaminated with test data. This means that public benchmarks used to assess LLMs are compromised, suggesting a performance gap between benchmark scores and actual capabilities. Ideally, a private holdout set could be used to accurately verify scores. Unfortunately, such datasets do not exist for most benchmarks, and post-hoc construction of sufficiently similar datasets is non-trivial. To address these issues, we introduce a systematic methodology for (i) retrospectively constructing a holdout dataset for a target dataset, (ii) demon","authors_text":"Cenny Wenner, Clement Neo, Esben Kran, Jacob Haimes, Jason Schreiber, Kunvar Thaman, Vassil Tashev","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-11T20:46:56Z","title":"Benchmark Inflation: Revealing LLM Performance Gaps Using Retro-Holdouts"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.09247","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:e876a4f13a14ce59f732e29c030929d00adb3bbfb7839050ff4681a9078fe339","target":"record","created_at":"2026-07-05T09:20:05Z","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":"11218bff621513be954c37ffef7677968450907c7af7b341d4bd343e8fe10a2c","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-11T20:46:56Z","title_canon_sha256":"fde39da6ddb8ad6edd159b5f82ab961732364b12bad98f856610673f0518841d"},"schema_version":"1.0","source":{"id":"2410.09247","kind":"arxiv","version":1}},"canonical_sha256":"f3b08f836bebca67cbfa29b05a51b5ec616b0eaa9da7f9bee3e83def67ee7f2c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f3b08f836bebca67cbfa29b05a51b5ec616b0eaa9da7f9bee3e83def67ee7f2c","first_computed_at":"2026-07-05T09:20:05.639362Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:20:05.639362Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nZGv2IphtdvU4LEwitjVLT4lKZFIbQ8IzeEUCgVP7sjfrH9NVKDyxVnu5qP5Chq2ZrsPY7ptn/pIiGG7D3YcCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:20:05.639757Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.09247","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e876a4f13a14ce59f732e29c030929d00adb3bbfb7839050ff4681a9078fe339","sha256:422db9398ce26f4b29c45e9b05bd784b4e7b017a22455c207f8cb226f1741aa5"],"state_sha256":"4531fe40163edb0c55e56501e7274ff389d199a1febbbbe5de71af339a844cd7"}