{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:AMQKACNMFXVHCOP3J362IDY2IM","short_pith_number":"pith:AMQKACNM","schema_version":"1.0","canonical_sha256":"0320a009ac2dea7139fb4efda40f1a430b6ee133302f06559c7f821b9416f13f","source":{"kind":"arxiv","id":"2507.06434","version":1},"attestation_state":"computed","paper":{"title":"Deprecating Benchmarks: Criteria and Framework","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CY","authors_text":"Ariel Gil, Ayrton San Joaquin, Leon Staufer, Rokas Gipi\\v{s}kis","submitted_at":"2025-07-08T22:29:06Z","abstract_excerpt":"As frontier artificial intelligence (AI) models rapidly advance, benchmarks are integral to comparing different models and measuring their progress in different task-specific domains. However, there is a lack of guidance on when and how benchmarks should be deprecated once they cease to effectively perform their purpose. This risks benchmark scores over-valuing model capabilities, or worse, obscuring capabilities and safety-washing. Based on a review of benchmarking practices, we propose criteria to decide when to fully or partially deprecate benchmarks, and a framework for deprecating benchma"},"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":"2507.06434","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2025-07-08T22:29:06Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"820fc1b2d581c09760d0d7af152129d949c1bd7795cbfdc0f1966544e5d001d6","abstract_canon_sha256":"4f8503cf0d33559983c933f9c5857e7051c54fe96bf880f6fc8ec9773f9ec8c1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:34:15.794636Z","signature_b64":"XAXbtlILMmGMiUL5wqNVXUxCZkMKwslhJgW39rnwMARAU//8GziepVzV248BTM2TdRmGjk9uqv3jDypGXsIMDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0320a009ac2dea7139fb4efda40f1a430b6ee133302f06559c7f821b9416f13f","last_reissued_at":"2026-07-05T11:34:15.794141Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:34:15.794141Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deprecating Benchmarks: Criteria and Framework","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CY","authors_text":"Ariel Gil, Ayrton San Joaquin, Leon Staufer, Rokas Gipi\\v{s}kis","submitted_at":"2025-07-08T22:29:06Z","abstract_excerpt":"As frontier artificial intelligence (AI) models rapidly advance, benchmarks are integral to comparing different models and measuring their progress in different task-specific domains. However, there is a lack of guidance on when and how benchmarks should be deprecated once they cease to effectively perform their purpose. This risks benchmark scores over-valuing model capabilities, or worse, obscuring capabilities and safety-washing. Based on a review of benchmarking practices, we propose criteria to decide when to fully or partially deprecate benchmarks, and a framework for deprecating benchma"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.06434","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/2507.06434/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":"2507.06434","created_at":"2026-07-05T11:34:15.794204+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.06434v1","created_at":"2026-07-05T11:34:15.794204+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.06434","created_at":"2026-07-05T11:34:15.794204+00:00"},{"alias_kind":"pith_short_12","alias_value":"AMQKACNMFXVH","created_at":"2026-07-05T11:34:15.794204+00:00"},{"alias_kind":"pith_short_16","alias_value":"AMQKACNMFXVHCOP3","created_at":"2026-07-05T11:34:15.794204+00:00"},{"alias_kind":"pith_short_8","alias_value":"AMQKACNM","created_at":"2026-07-05T11:34:15.794204+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14164","citing_title":"Unsteady Metrics and Benchmarking Cultures of AI Model Builders","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AMQKACNMFXVHCOP3J362IDY2IM","json":"https://pith.science/pith/AMQKACNMFXVHCOP3J362IDY2IM.json","graph_json":"https://pith.science/api/pith-number/AMQKACNMFXVHCOP3J362IDY2IM/graph.json","events_json":"https://pith.science/api/pith-number/AMQKACNMFXVHCOP3J362IDY2IM/events.json","paper":"https://pith.science/paper/AMQKACNM"},"agent_actions":{"view_html":"https://pith.science/pith/AMQKACNMFXVHCOP3J362IDY2IM","download_json":"https://pith.science/pith/AMQKACNMFXVHCOP3J362IDY2IM.json","view_paper":"https://pith.science/paper/AMQKACNM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.06434&json=true","fetch_graph":"https://pith.science/api/pith-number/AMQKACNMFXVHCOP3J362IDY2IM/graph.json","fetch_events":"https://pith.science/api/pith-number/AMQKACNMFXVHCOP3J362IDY2IM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AMQKACNMFXVHCOP3J362IDY2IM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AMQKACNMFXVHCOP3J362IDY2IM/action/storage_attestation","attest_author":"https://pith.science/pith/AMQKACNMFXVHCOP3J362IDY2IM/action/author_attestation","sign_citation":"https://pith.science/pith/AMQKACNMFXVHCOP3J362IDY2IM/action/citation_signature","submit_replication":"https://pith.science/pith/AMQKACNMFXVHCOP3J362IDY2IM/action/replication_record"}},"created_at":"2026-07-05T11:34:15.794204+00:00","updated_at":"2026-07-05T11:34:15.794204+00:00"}