{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KAD6FGDO5USZZRKLWSNWFEVJZ3","short_pith_number":"pith:KAD6FGDO","schema_version":"1.0","canonical_sha256":"5007e2986eed259cc54bb49b6292a9cee30497edea383f4d243bcd3c1f50bb83","source":{"kind":"arxiv","id":"2606.26158","version":1},"attestation_state":"computed","paper":{"title":"Life After Benchmark Saturation: A Case Study of CORE-Bench","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","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","submitted_at":"2026-06-23T22:30:44Z","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"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.26158","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-23T22:30:44Z","cross_cats_sorted":[],"title_canon_sha256":"3bb456adab8d98ca19952ae70b2a2dc8df3a70bc863e9ac382a8fd0a05ef28b0","abstract_canon_sha256":"f908f58f24a3c7c483c221724a96948ea1805cf201963f0ae28e699feb621548"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-26T00:15:27.563133Z","signature_b64":"w8My47X72KcuAAWtyhsUuIfxaBX5LJkqEsbBKcEsx73YsBoFxlmu5jq9t6NB7Ypryo/GrytjEaaWDFIlNMfpAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5007e2986eed259cc54bb49b6292a9cee30497edea383f4d243bcd3c1f50bb83","last_reissued_at":"2026-06-26T00:15:27.562610Z","signature_status":"signed_v1","first_computed_at":"2026-06-26T00:15:27.562610Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Life After Benchmark Saturation: A Case Study of CORE-Bench","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","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","submitted_at":"2026-06-23T22:30:44Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26158","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.26158/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.26158","created_at":"2026-06-26T00:15:27.562679+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.26158v1","created_at":"2026-06-26T00:15:27.562679+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26158","created_at":"2026-06-26T00:15:27.562679+00:00"},{"alias_kind":"pith_short_12","alias_value":"KAD6FGDO5USZ","created_at":"2026-06-26T00:15:27.562679+00:00"},{"alias_kind":"pith_short_16","alias_value":"KAD6FGDO5USZZRKL","created_at":"2026-06-26T00:15:27.562679+00:00"},{"alias_kind":"pith_short_8","alias_value":"KAD6FGDO","created_at":"2026-06-26T00:15:27.562679+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KAD6FGDO5USZZRKLWSNWFEVJZ3","json":"https://pith.science/pith/KAD6FGDO5USZZRKLWSNWFEVJZ3.json","graph_json":"https://pith.science/api/pith-number/KAD6FGDO5USZZRKLWSNWFEVJZ3/graph.json","events_json":"https://pith.science/api/pith-number/KAD6FGDO5USZZRKLWSNWFEVJZ3/events.json","paper":"https://pith.science/paper/KAD6FGDO"},"agent_actions":{"view_html":"https://pith.science/pith/KAD6FGDO5USZZRKLWSNWFEVJZ3","download_json":"https://pith.science/pith/KAD6FGDO5USZZRKLWSNWFEVJZ3.json","view_paper":"https://pith.science/paper/KAD6FGDO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.26158&json=true","fetch_graph":"https://pith.science/api/pith-number/KAD6FGDO5USZZRKLWSNWFEVJZ3/graph.json","fetch_events":"https://pith.science/api/pith-number/KAD6FGDO5USZZRKLWSNWFEVJZ3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KAD6FGDO5USZZRKLWSNWFEVJZ3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KAD6FGDO5USZZRKLWSNWFEVJZ3/action/storage_attestation","attest_author":"https://pith.science/pith/KAD6FGDO5USZZRKLWSNWFEVJZ3/action/author_attestation","sign_citation":"https://pith.science/pith/KAD6FGDO5USZZRKLWSNWFEVJZ3/action/citation_signature","submit_replication":"https://pith.science/pith/KAD6FGDO5USZZRKLWSNWFEVJZ3/action/replication_record"}},"created_at":"2026-06-26T00:15:27.562679+00:00","updated_at":"2026-06-26T00:15:27.562679+00:00"}