{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:KAD6FGDO5USZZRKLWSNWFEVJZ3","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":"f908f58f24a3c7c483c221724a96948ea1805cf201963f0ae28e699feb621548","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-23T22:30:44Z","title_canon_sha256":"3bb456adab8d98ca19952ae70b2a2dc8df3a70bc863e9ac382a8fd0a05ef28b0"},"schema_version":"1.0","source":{"id":"2606.26158","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.26158","created_at":"2026-06-26T00:15:27Z"},{"alias_kind":"arxiv_version","alias_value":"2606.26158v1","created_at":"2026-06-26T00:15:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26158","created_at":"2026-06-26T00:15:27Z"},{"alias_kind":"pith_short_12","alias_value":"KAD6FGDO5USZ","created_at":"2026-06-26T00:15:27Z"},{"alias_kind":"pith_short_16","alias_value":"KAD6FGDO5USZZRKL","created_at":"2026-06-26T00:15:27Z"},{"alias_kind":"pith_short_8","alias_value":"KAD6FGDO","created_at":"2026-06-26T00:15:27Z"}],"graph_snapshots":[{"event_id":"sha256:6a080dfd86d4b135e80a2300a93b27be3b676be1dd4ebf014b86c67f9872afa0","target":"graph","created_at":"2026-06-26T00:15:27Z","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/2606.26158/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"When a benchmark's accuracy saturates, it is often retired and replaced with a more challenging version. We show that this approach privileges accuracy and misses the opportunity to study six other key dimensions of agent performance: construct validity issues such as shortcuts, out-of-distribution generalizability, efficiency, reliability, the relative importance of the model versus the scaffold, and uplift from human-agent collaboration. We use CORE-Bench Hard, a benchmark for computational reproducibility of scientific code, as a case study to demonstrate that measuring agents along these d","authors_text":"Abhishek Shetty, Arvind Narayanan, Derrick Chan-Sew, Kangheng Liu, Matilda Orona, Nitya Nadgir, Peter Kirgis, Rumi Nakagawa, Saiteja Utpala, Sayash Kapoor, Stephan Rabanser, Tilman Bayer, Yue Ling, Zachary S. Siegel","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-23T22:30:44Z","title":"Life After Benchmark Saturation: A Case Study of CORE-Bench"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26158","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:f5ac04d57cc1ec06d618390d5e7022b13d4bd20eeca033d4e01009e32359c9ec","target":"record","created_at":"2026-06-26T00:15:27Z","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":"f908f58f24a3c7c483c221724a96948ea1805cf201963f0ae28e699feb621548","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-23T22:30:44Z","title_canon_sha256":"3bb456adab8d98ca19952ae70b2a2dc8df3a70bc863e9ac382a8fd0a05ef28b0"},"schema_version":"1.0","source":{"id":"2606.26158","kind":"arxiv","version":1}},"canonical_sha256":"5007e2986eed259cc54bb49b6292a9cee30497edea383f4d243bcd3c1f50bb83","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5007e2986eed259cc54bb49b6292a9cee30497edea383f4d243bcd3c1f50bb83","first_computed_at":"2026-06-26T00:15:27.562610Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-26T00:15:27.562610Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"w8My47X72KcuAAWtyhsUuIfxaBX5LJkqEsbBKcEsx73YsBoFxlmu5jq9t6NB7Ypryo/GrytjEaaWDFIlNMfpAQ==","signature_status":"signed_v1","signed_at":"2026-06-26T00:15:27.563133Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.26158","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f5ac04d57cc1ec06d618390d5e7022b13d4bd20eeca033d4e01009e32359c9ec","sha256:6a080dfd86d4b135e80a2300a93b27be3b676be1dd4ebf014b86c67f9872afa0"],"state_sha256":"54b488699335f15676251c6792ce8c4a9ecb6aea2b04a3d534137f03a04122ee"}