{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:6VLOCLUDCQYJT35G5BXKW2BUY5","short_pith_number":"pith:6VLOCLUD","canonical_record":{"source":{"id":"2411.12990","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-11-20T02:38:24Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"918f4c44fa8cc35a1eeebe409f6760f34fc9da85f538ead666f83ef08a709539","abstract_canon_sha256":"cffd1f5aaaf8ef476580fd1cad65a8381d758044b3de8015967f380d4fa58a4c"},"schema_version":"1.0"},"canonical_sha256":"f556e12e83143099efa6e86eab6834c74868c9ceeef97f45879ea5adcca3b034","source":{"kind":"arxiv","id":"2411.12990","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.12990","created_at":"2026-07-05T09:38:03Z"},{"alias_kind":"arxiv_version","alias_value":"2411.12990v1","created_at":"2026-07-05T09:38:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.12990","created_at":"2026-07-05T09:38:03Z"},{"alias_kind":"pith_short_12","alias_value":"6VLOCLUDCQYJ","created_at":"2026-07-05T09:38:03Z"},{"alias_kind":"pith_short_16","alias_value":"6VLOCLUDCQYJT35G","created_at":"2026-07-05T09:38:03Z"},{"alias_kind":"pith_short_8","alias_value":"6VLOCLUD","created_at":"2026-07-05T09:38:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:6VLOCLUDCQYJT35G5BXKW2BUY5","target":"record","payload":{"canonical_record":{"source":{"id":"2411.12990","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-11-20T02:38:24Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"918f4c44fa8cc35a1eeebe409f6760f34fc9da85f538ead666f83ef08a709539","abstract_canon_sha256":"cffd1f5aaaf8ef476580fd1cad65a8381d758044b3de8015967f380d4fa58a4c"},"schema_version":"1.0"},"canonical_sha256":"f556e12e83143099efa6e86eab6834c74868c9ceeef97f45879ea5adcca3b034","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:38:03.337866Z","signature_b64":"I2RerVPH3khax5HTlPNA4ecXImQG2+rHsoZV+ow5rfJhWdHUfZ3lQlONaApzrpSdr0M7y6Md4b6ct2qL18esAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f556e12e83143099efa6e86eab6834c74868c9ceeef97f45879ea5adcca3b034","last_reissued_at":"2026-07-05T09:38:03.337373Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:38:03.337373Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.12990","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:38:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PS+Xbipl1Z8xJFcO6Bw8x4TCk/60G4350SwIf8pF64dptCYa5NmsS0oOU6ZyizCbnS8USWHUbezQK3iWhty2BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T18:38:36.351678Z"},"content_sha256":"ce425840941b8ce64b7b236cfb1deb4a84f3a6122d9834e1514573d7621c3240","schema_version":"1.0","event_id":"sha256:ce425840941b8ce64b7b236cfb1deb4a84f3a6122d9834e1514573d7621c3240"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:6VLOCLUDCQYJT35G5BXKW2BUY5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"BetterBench: Assessing AI Benchmarks, Uncovering Issues, and Establishing Best Practices","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Amelia Hardy, Anka Reuel, Chandler Smith, Malcolm Hardy, Max Lamparth, Mykel J. Kochenderfer","submitted_at":"2024-11-20T02:38:24Z","abstract_excerpt":"AI models are increasingly prevalent in high-stakes environments, necessitating thorough assessment of their capabilities and risks. Benchmarks are popular for measuring these attributes and for comparing model performance, tracking progress, and identifying weaknesses in foundation and non-foundation models. They can inform model selection for downstream tasks and influence policy initiatives. However, not all benchmarks are the same: their quality depends on their design and usability. In this paper, we develop an assessment framework considering 46 best practices across an AI benchmark's li"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.12990","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/2411.12990/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:38:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"euHTYFsssD6Aij6rBiyTr0trk74TksoAFTwXvkoCqLeLYddX/l9IjwITb44wKQodAY+Wb32aXooAG64Ix+fzBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T18:38:36.352191Z"},"content_sha256":"30923985224e7320a2a16de4b9e79ef8c1fd63ca647269b3bb0a6d92c1d11a14","schema_version":"1.0","event_id":"sha256:30923985224e7320a2a16de4b9e79ef8c1fd63ca647269b3bb0a6d92c1d11a14"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6VLOCLUDCQYJT35G5BXKW2BUY5/bundle.json","state_url":"https://pith.science/pith/6VLOCLUDCQYJT35G5BXKW2BUY5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6VLOCLUDCQYJT35G5BXKW2BUY5/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T18:38:36Z","links":{"resolver":"https://pith.science/pith/6VLOCLUDCQYJT35G5BXKW2BUY5","bundle":"https://pith.science/pith/6VLOCLUDCQYJT35G5BXKW2BUY5/bundle.json","state":"https://pith.science/pith/6VLOCLUDCQYJT35G5BXKW2BUY5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6VLOCLUDCQYJT35G5BXKW2BUY5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:6VLOCLUDCQYJT35G5BXKW2BUY5","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":"cffd1f5aaaf8ef476580fd1cad65a8381d758044b3de8015967f380d4fa58a4c","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-11-20T02:38:24Z","title_canon_sha256":"918f4c44fa8cc35a1eeebe409f6760f34fc9da85f538ead666f83ef08a709539"},"schema_version":"1.0","source":{"id":"2411.12990","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.12990","created_at":"2026-07-05T09:38:03Z"},{"alias_kind":"arxiv_version","alias_value":"2411.12990v1","created_at":"2026-07-05T09:38:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.12990","created_at":"2026-07-05T09:38:03Z"},{"alias_kind":"pith_short_12","alias_value":"6VLOCLUDCQYJ","created_at":"2026-07-05T09:38:03Z"},{"alias_kind":"pith_short_16","alias_value":"6VLOCLUDCQYJT35G","created_at":"2026-07-05T09:38:03Z"},{"alias_kind":"pith_short_8","alias_value":"6VLOCLUD","created_at":"2026-07-05T09:38:03Z"}],"graph_snapshots":[{"event_id":"sha256:30923985224e7320a2a16de4b9e79ef8c1fd63ca647269b3bb0a6d92c1d11a14","target":"graph","created_at":"2026-07-05T09:38:03Z","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/2411.12990/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"AI models are increasingly prevalent in high-stakes environments, necessitating thorough assessment of their capabilities and risks. Benchmarks are popular for measuring these attributes and for comparing model performance, tracking progress, and identifying weaknesses in foundation and non-foundation models. They can inform model selection for downstream tasks and influence policy initiatives. However, not all benchmarks are the same: their quality depends on their design and usability. In this paper, we develop an assessment framework considering 46 best practices across an AI benchmark's li","authors_text":"Amelia Hardy, Anka Reuel, Chandler Smith, Malcolm Hardy, Max Lamparth, Mykel J. Kochenderfer","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-11-20T02:38:24Z","title":"BetterBench: Assessing AI Benchmarks, Uncovering Issues, and Establishing Best Practices"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.12990","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:ce425840941b8ce64b7b236cfb1deb4a84f3a6122d9834e1514573d7621c3240","target":"record","created_at":"2026-07-05T09:38:03Z","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":"cffd1f5aaaf8ef476580fd1cad65a8381d758044b3de8015967f380d4fa58a4c","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-11-20T02:38:24Z","title_canon_sha256":"918f4c44fa8cc35a1eeebe409f6760f34fc9da85f538ead666f83ef08a709539"},"schema_version":"1.0","source":{"id":"2411.12990","kind":"arxiv","version":1}},"canonical_sha256":"f556e12e83143099efa6e86eab6834c74868c9ceeef97f45879ea5adcca3b034","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f556e12e83143099efa6e86eab6834c74868c9ceeef97f45879ea5adcca3b034","first_computed_at":"2026-07-05T09:38:03.337373Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:38:03.337373Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"I2RerVPH3khax5HTlPNA4ecXImQG2+rHsoZV+ow5rfJhWdHUfZ3lQlONaApzrpSdr0M7y6Md4b6ct2qL18esAg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:38:03.337866Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.12990","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ce425840941b8ce64b7b236cfb1deb4a84f3a6122d9834e1514573d7621c3240","sha256:30923985224e7320a2a16de4b9e79ef8c1fd63ca647269b3bb0a6d92c1d11a14"],"state_sha256":"83841ba1d3f395b42533768d8a16174634a2ed6a8189a434ed9467de7827600b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5+ZCMPi2EpGirkfPJle0QppYJf/tCKKsql+zg9EF4jWsK9VV6oCKDuAugMP+mEJaq8zSRy52KD0XSG0ZMt5VDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T18:38:36.356096Z","bundle_sha256":"521560d8fe707e6aeeb0ef39e980adfe431ac576a4e64deca2ac49c5739c1275"}}