{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:WTK4ZQSQQGJFTTY6YJ6CPVEOPW","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":"bfacf113081f1f848115be1bb01c43cde17fcedc9273619627f9016fcbf04a59","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-19T09:31:50Z","title_canon_sha256":"beda9d7076d11ae2db6ba54c05f2a62b621deab4955cf1da881034896043fa98"},"schema_version":"1.0","source":{"id":"2502.13576","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.13576","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"arxiv_version","alias_value":"2502.13576v2","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.13576","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"pith_short_12","alias_value":"WTK4ZQSQQGJF","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"pith_short_16","alias_value":"WTK4ZQSQQGJFTTY6","created_at":"2026-07-05T11:13:20Z"},{"alias_kind":"pith_short_8","alias_value":"WTK4ZQSQ","created_at":"2026-07-05T11:13:20Z"}],"graph_snapshots":[{"event_id":"sha256:26e76639534feadba731dc7d4f5ed964ac0d6d9059cbade972e5b43c5cb63be1","target":"graph","created_at":"2026-07-05T11:13:20Z","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/2502.13576/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluating models on large benchmarks is very resource-intensive, especially during the period of rapid model evolution. Existing efficient evaluation methods estimate the performance of target models by testing them only on a small and static coreset of the benchmark, which is derived from the publicly available evaluation results of source models. These methods rely on the assumption that target models have high prediction consistency with source models. However, we demonstrate that it doesn't generalize well in practice. To alleviate the inconsistency issue, we present TailoredBench, a meth","authors_text":"Boyuan Pan, Chuyi Tan, Jiayi Shi, Kan Li, Peiwen Yuan, Shaoxiong Feng, Xinglin Wang, Yao Hu, Yiwei Li, Yueqi Zhang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-19T09:31:50Z","title":"Beyond One-Size-Fits-All: Tailored Benchmarks for Efficient Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.13576","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":"sha256:0affff0b50a128b6703804e61771bf6edfd0f990f701f1193a69358b40f5290b","target":"record","created_at":"2026-07-05T11:13:20Z","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":"bfacf113081f1f848115be1bb01c43cde17fcedc9273619627f9016fcbf04a59","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-19T09:31:50Z","title_canon_sha256":"beda9d7076d11ae2db6ba54c05f2a62b621deab4955cf1da881034896043fa98"},"schema_version":"1.0","source":{"id":"2502.13576","kind":"arxiv","version":2}},"canonical_sha256":"b4d5ccc250819259cf1ec27c27d48e7d8d0da2cdea9dfdab344379a0d7e11530","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b4d5ccc250819259cf1ec27c27d48e7d8d0da2cdea9dfdab344379a0d7e11530","first_computed_at":"2026-07-05T11:13:20.808804Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:20.808804Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cHvhpiaZvF7CLrZ3aL9I9sIAVOEBKjkiTsxIJg20H0HvAeRLL/cOZkTSQyO9UXO/HnWSUmMP/t4sPxsR4UiqDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:20.809275Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.13576","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0affff0b50a128b6703804e61771bf6edfd0f990f701f1193a69358b40f5290b","sha256:26e76639534feadba731dc7d4f5ed964ac0d6d9059cbade972e5b43c5cb63be1"],"state_sha256":"1fd8514a43d7c03aa743956acb41df9fcca62ff2122f7505c29cd33fe2f695a3"}