{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:7RFBDVMFVMACFZQCJJYONTO2GW","short_pith_number":"pith:7RFBDVMF","canonical_record":{"source":{"id":"2008.02577","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2020-08-06T11:14:37Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9a0a1ea8c5e004924e6ec57f82b3fa4a81310e38c8b83d1ef8db5bbd139fd8fb","abstract_canon_sha256":"19276e4ec989803bbab6f3c1da4b361c71b945f587dd76e9014878f7fe128407"},"schema_version":"1.0"},"canonical_sha256":"fc4a11d585ab0022e6024a70e6cdda35be9eac4ebf9c8d0c4830c3117b648870","source":{"kind":"arxiv","id":"2008.02577","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2008.02577","created_at":"2026-07-05T03:29:45Z"},{"alias_kind":"arxiv_version","alias_value":"2008.02577v2","created_at":"2026-07-05T03:29:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.02577","created_at":"2026-07-05T03:29:45Z"},{"alias_kind":"pith_short_12","alias_value":"7RFBDVMFVMAC","created_at":"2026-07-05T03:29:45Z"},{"alias_kind":"pith_short_16","alias_value":"7RFBDVMFVMACFZQC","created_at":"2026-07-05T03:29:45Z"},{"alias_kind":"pith_short_8","alias_value":"7RFBDVMF","created_at":"2026-07-05T03:29:45Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:7RFBDVMFVMACFZQCJJYONTO2GW","target":"record","payload":{"canonical_record":{"source":{"id":"2008.02577","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2020-08-06T11:14:37Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9a0a1ea8c5e004924e6ec57f82b3fa4a81310e38c8b83d1ef8db5bbd139fd8fb","abstract_canon_sha256":"19276e4ec989803bbab6f3c1da4b361c71b945f587dd76e9014878f7fe128407"},"schema_version":"1.0"},"canonical_sha256":"fc4a11d585ab0022e6024a70e6cdda35be9eac4ebf9c8d0c4830c3117b648870","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:29:45.659115Z","signature_b64":"RSqZ/uCYf7FiBlkJ3t0AUQi93ql46IQhARXxtorj9m3frcisIj450Cm0biFzmztwP8qqOe+6qsNwDao7uqbVCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fc4a11d585ab0022e6024a70e6cdda35be9eac4ebf9c8d0c4830c3117b648870","last_reissued_at":"2026-07-05T03:29:45.658583Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:29:45.658583Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2008.02577","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:29:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aseQX5XqcyTgo+NqAGsmODrORiC0PcFruOqZCepTSCyX2h26KFzpa0WpPf8iqrbxyIvN4tih/DeG8HPIF8rxAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T21:33:42.537950Z"},"content_sha256":"1fb2aa1b646ac23310a7a5d2d2a82d739fc501ad129fa38d4b758efba65d4ecb","schema_version":"1.0","event_id":"sha256:1fb2aa1b646ac23310a7a5d2d2a82d739fc501ad129fa38d4b758efba65d4ecb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:7RFBDVMFVMACFZQCJJYONTO2GW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A critical analysis of metrics used for measuring progress in artificial intelligence","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Georg Dorffner, Kathrin Blagec, Matthias Samwald, Milad Moradi","submitted_at":"2020-08-06T11:14:37Z","abstract_excerpt":"Comparing model performances on benchmark datasets is an integral part of measuring and driving progress in artificial intelligence. A model's performance on a benchmark dataset is commonly assessed based on a single or a small set of performance metrics. While this enables quick comparisons, it may entail the risk of inadequately reflecting model performance if the metric does not sufficiently cover all performance characteristics. It is unknown to what extent this might impact benchmarking efforts.\n  To address this question, we analysed the current landscape of performance metrics based on "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.02577","kind":"arxiv","version":2},"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/2008.02577/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":"2026-07-05T03:29:45Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EWr9EIHMSqTxHJVfpjZqeTNT1jotFK0uKsW+t10yDD/mAub3Ip6mjXSyiuJKI7AIUbxMLYTIss70ysdaX19IBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T21:33:42.538555Z"},"content_sha256":"c2118de75f7d93ea13c5d4be1663233ba14697fe4fb8f2103b4599acca0517e3","schema_version":"1.0","event_id":"sha256:c2118de75f7d93ea13c5d4be1663233ba14697fe4fb8f2103b4599acca0517e3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7RFBDVMFVMACFZQCJJYONTO2GW/bundle.json","state_url":"https://pith.science/pith/7RFBDVMFVMACFZQCJJYONTO2GW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7RFBDVMFVMACFZQCJJYONTO2GW/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":"2026-08-05T21:33:42Z","links":{"resolver":"https://pith.science/pith/7RFBDVMFVMACFZQCJJYONTO2GW","bundle":"https://pith.science/pith/7RFBDVMFVMACFZQCJJYONTO2GW/bundle.json","state":"https://pith.science/pith/7RFBDVMFVMACFZQCJJYONTO2GW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7RFBDVMFVMACFZQCJJYONTO2GW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:7RFBDVMFVMACFZQCJJYONTO2GW","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":"19276e4ec989803bbab6f3c1da4b361c71b945f587dd76e9014878f7fe128407","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2020-08-06T11:14:37Z","title_canon_sha256":"9a0a1ea8c5e004924e6ec57f82b3fa4a81310e38c8b83d1ef8db5bbd139fd8fb"},"schema_version":"1.0","source":{"id":"2008.02577","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2008.02577","created_at":"2026-07-05T03:29:45Z"},{"alias_kind":"arxiv_version","alias_value":"2008.02577v2","created_at":"2026-07-05T03:29:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2008.02577","created_at":"2026-07-05T03:29:45Z"},{"alias_kind":"pith_short_12","alias_value":"7RFBDVMFVMAC","created_at":"2026-07-05T03:29:45Z"},{"alias_kind":"pith_short_16","alias_value":"7RFBDVMFVMACFZQC","created_at":"2026-07-05T03:29:45Z"},{"alias_kind":"pith_short_8","alias_value":"7RFBDVMF","created_at":"2026-07-05T03:29:45Z"}],"graph_snapshots":[{"event_id":"sha256:c2118de75f7d93ea13c5d4be1663233ba14697fe4fb8f2103b4599acca0517e3","target":"graph","created_at":"2026-07-05T03:29:45Z","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/2008.02577/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Comparing model performances on benchmark datasets is an integral part of measuring and driving progress in artificial intelligence. A model's performance on a benchmark dataset is commonly assessed based on a single or a small set of performance metrics. While this enables quick comparisons, it may entail the risk of inadequately reflecting model performance if the metric does not sufficiently cover all performance characteristics. It is unknown to what extent this might impact benchmarking efforts.\n  To address this question, we analysed the current landscape of performance metrics based on ","authors_text":"Georg Dorffner, Kathrin Blagec, Matthias Samwald, Milad Moradi","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2020-08-06T11:14:37Z","title":"A critical analysis of metrics used for measuring progress in artificial intelligence"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2008.02577","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":"sha256:1fb2aa1b646ac23310a7a5d2d2a82d739fc501ad129fa38d4b758efba65d4ecb","target":"record","created_at":"2026-07-05T03:29:45Z","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":"19276e4ec989803bbab6f3c1da4b361c71b945f587dd76e9014878f7fe128407","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2020-08-06T11:14:37Z","title_canon_sha256":"9a0a1ea8c5e004924e6ec57f82b3fa4a81310e38c8b83d1ef8db5bbd139fd8fb"},"schema_version":"1.0","source":{"id":"2008.02577","kind":"arxiv","version":2}},"canonical_sha256":"fc4a11d585ab0022e6024a70e6cdda35be9eac4ebf9c8d0c4830c3117b648870","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fc4a11d585ab0022e6024a70e6cdda35be9eac4ebf9c8d0c4830c3117b648870","first_computed_at":"2026-07-05T03:29:45.658583Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:29:45.658583Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RSqZ/uCYf7FiBlkJ3t0AUQi93ql46IQhARXxtorj9m3frcisIj450Cm0biFzmztwP8qqOe+6qsNwDao7uqbVCg==","signature_status":"signed_v1","signed_at":"2026-07-05T03:29:45.659115Z","signed_message":"canonical_sha256_bytes"},"source_id":"2008.02577","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1fb2aa1b646ac23310a7a5d2d2a82d739fc501ad129fa38d4b758efba65d4ecb","sha256:c2118de75f7d93ea13c5d4be1663233ba14697fe4fb8f2103b4599acca0517e3"],"state_sha256":"acc825127a9ddac3c73bff4bde33deb430883550c844b3b8cc14355486b8a156"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nhvYA5N6sr1TVaDfrbiATM6yy22O3r/1kIvQ922D82LmqET2/2mv4CLFpLHkHIq+x3cyAHlfgMMkf0d0UIsmAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T21:33:42.542572Z","bundle_sha256":"c2b5a4287f56538f2f30955e6b4987cc738d58705e36ea2eb00b54c514e77955"}}