{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:EWVFL64ZNUHHM4ZJZRZWNZ6FDK","short_pith_number":"pith:EWVFL64Z","canonical_record":{"source":{"id":"1806.02643","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-07T12:38:19Z","cross_cats_sorted":["cs.GT","stat.ML"],"title_canon_sha256":"7add2680a4f4ce1cbec7618854eaab3c973ffa1443ee9ac62057b38d89c93597","abstract_canon_sha256":"1b8ef308079e9c3c67de3268faedfac555440ec0433c755095379c36a18cf847"},"schema_version":"1.0"},"canonical_sha256":"25aa55fb996d0e767329cc7366e7c51aa42b326e1560a3dc0f4e2f106adbc925","source":{"kind":"arxiv","id":"1806.02643","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.02643","created_at":"2026-05-18T00:01:53Z"},{"alias_kind":"arxiv_version","alias_value":"1806.02643v2","created_at":"2026-05-18T00:01:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.02643","created_at":"2026-05-18T00:01:53Z"},{"alias_kind":"pith_short_12","alias_value":"EWVFL64ZNUHH","created_at":"2026-05-18T12:32:22Z"},{"alias_kind":"pith_short_16","alias_value":"EWVFL64ZNUHHM4ZJ","created_at":"2026-05-18T12:32:22Z"},{"alias_kind":"pith_short_8","alias_value":"EWVFL64Z","created_at":"2026-05-18T12:32:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:EWVFL64ZNUHHM4ZJZRZWNZ6FDK","target":"record","payload":{"canonical_record":{"source":{"id":"1806.02643","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-07T12:38:19Z","cross_cats_sorted":["cs.GT","stat.ML"],"title_canon_sha256":"7add2680a4f4ce1cbec7618854eaab3c973ffa1443ee9ac62057b38d89c93597","abstract_canon_sha256":"1b8ef308079e9c3c67de3268faedfac555440ec0433c755095379c36a18cf847"},"schema_version":"1.0"},"canonical_sha256":"25aa55fb996d0e767329cc7366e7c51aa42b326e1560a3dc0f4e2f106adbc925","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:01:53.171867Z","signature_b64":"qqhNSz241r/96xNy6llatuAiXZl80QEJ4Fxl+BRZ02z4ApIF9XsPyhaeHMjVoNntGR/nsT0itKE9+R/rQxGFDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"25aa55fb996d0e767329cc7366e7c51aa42b326e1560a3dc0f4e2f106adbc925","last_reissued_at":"2026-05-18T00:01:53.171407Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:01:53.171407Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1806.02643","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":"2026-05-18T00:01:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nr59t9Z1HUhxFxdljsJxBPGk4ch6MDACB5CclZx3xqiqPkdhGKnvAI843hVMHb4pu4PrrWnRn+56PcTxZomMCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T18:23:41.315634Z"},"content_sha256":"9e7f26257001a7d471ba3fff56a8a085b3a577459d3044dbb7ffe51edf852bfb","schema_version":"1.0","event_id":"sha256:9e7f26257001a7d471ba3fff56a8a085b3a577459d3044dbb7ffe51edf852bfb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:EWVFL64ZNUHHM4ZJZRZWNZ6FDK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Re-evaluating Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.GT","stat.ML"],"primary_cat":"cs.LG","authors_text":"David Balduzzi, Julien Perolat, Karl Tuyls, Thore Graepel","submitted_at":"2018-06-07T12:38:19Z","abstract_excerpt":"Progress in machine learning is measured by careful evaluation on problems of outstanding common interest. However, the proliferation of benchmark suites and environments, adversarial attacks, and other complications has diluted the basic evaluation model by overwhelming researchers with choices. Deliberate or accidental cherry picking is increasingly likely, and designing well-balanced evaluation suites requires increasing effort. In this paper we take a step back and propose Nash averaging. The approach builds on a detailed analysis of the algebraic structure of evaluation in two basic scena"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.02643","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":""},"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":"2026-05-18T00:01:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tM5smtcLaTZBXMLQKAW2+Az/5eOjbxjsljLnMdnrQ4WoLqiKRQWlYwljJNvZr6k98cA8Z9Fe3ZPVzchhhYxOBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T18:23:41.315981Z"},"content_sha256":"7ae22f93e9252de881fe26eb369ab66e9b1b9e23d2df92df6a7f41b89d828634","schema_version":"1.0","event_id":"sha256:7ae22f93e9252de881fe26eb369ab66e9b1b9e23d2df92df6a7f41b89d828634"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EWVFL64ZNUHHM4ZJZRZWNZ6FDK/bundle.json","state_url":"https://pith.science/pith/EWVFL64ZNUHHM4ZJZRZWNZ6FDK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EWVFL64ZNUHHM4ZJZRZWNZ6FDK/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":"2026-06-29T18:23:41Z","links":{"resolver":"https://pith.science/pith/EWVFL64ZNUHHM4ZJZRZWNZ6FDK","bundle":"https://pith.science/pith/EWVFL64ZNUHHM4ZJZRZWNZ6FDK/bundle.json","state":"https://pith.science/pith/EWVFL64ZNUHHM4ZJZRZWNZ6FDK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EWVFL64ZNUHHM4ZJZRZWNZ6FDK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:EWVFL64ZNUHHM4ZJZRZWNZ6FDK","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":"1b8ef308079e9c3c67de3268faedfac555440ec0433c755095379c36a18cf847","cross_cats_sorted":["cs.GT","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-07T12:38:19Z","title_canon_sha256":"7add2680a4f4ce1cbec7618854eaab3c973ffa1443ee9ac62057b38d89c93597"},"schema_version":"1.0","source":{"id":"1806.02643","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.02643","created_at":"2026-05-18T00:01:53Z"},{"alias_kind":"arxiv_version","alias_value":"1806.02643v2","created_at":"2026-05-18T00:01:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.02643","created_at":"2026-05-18T00:01:53Z"},{"alias_kind":"pith_short_12","alias_value":"EWVFL64ZNUHH","created_at":"2026-05-18T12:32:22Z"},{"alias_kind":"pith_short_16","alias_value":"EWVFL64ZNUHHM4ZJ","created_at":"2026-05-18T12:32:22Z"},{"alias_kind":"pith_short_8","alias_value":"EWVFL64Z","created_at":"2026-05-18T12:32:22Z"}],"graph_snapshots":[{"event_id":"sha256:7ae22f93e9252de881fe26eb369ab66e9b1b9e23d2df92df6a7f41b89d828634","target":"graph","created_at":"2026-05-18T00:01:53Z","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"},"paper":{"abstract_excerpt":"Progress in machine learning is measured by careful evaluation on problems of outstanding common interest. However, the proliferation of benchmark suites and environments, adversarial attacks, and other complications has diluted the basic evaluation model by overwhelming researchers with choices. Deliberate or accidental cherry picking is increasingly likely, and designing well-balanced evaluation suites requires increasing effort. In this paper we take a step back and propose Nash averaging. The approach builds on a detailed analysis of the algebraic structure of evaluation in two basic scena","authors_text":"David Balduzzi, Julien Perolat, Karl Tuyls, Thore Graepel","cross_cats":["cs.GT","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-07T12:38:19Z","title":"Re-evaluating Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.02643","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":"sha256:9e7f26257001a7d471ba3fff56a8a085b3a577459d3044dbb7ffe51edf852bfb","target":"record","created_at":"2026-05-18T00:01:53Z","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":"1b8ef308079e9c3c67de3268faedfac555440ec0433c755095379c36a18cf847","cross_cats_sorted":["cs.GT","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-07T12:38:19Z","title_canon_sha256":"7add2680a4f4ce1cbec7618854eaab3c973ffa1443ee9ac62057b38d89c93597"},"schema_version":"1.0","source":{"id":"1806.02643","kind":"arxiv","version":2}},"canonical_sha256":"25aa55fb996d0e767329cc7366e7c51aa42b326e1560a3dc0f4e2f106adbc925","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"25aa55fb996d0e767329cc7366e7c51aa42b326e1560a3dc0f4e2f106adbc925","first_computed_at":"2026-05-18T00:01:53.171407Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:01:53.171407Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qqhNSz241r/96xNy6llatuAiXZl80QEJ4Fxl+BRZ02z4ApIF9XsPyhaeHMjVoNntGR/nsT0itKE9+R/rQxGFDA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:01:53.171867Z","signed_message":"canonical_sha256_bytes"},"source_id":"1806.02643","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9e7f26257001a7d471ba3fff56a8a085b3a577459d3044dbb7ffe51edf852bfb","sha256:7ae22f93e9252de881fe26eb369ab66e9b1b9e23d2df92df6a7f41b89d828634"],"state_sha256":"4c2b91d627ebd291e1d396bc24e211817c513238eb94652c0bbeabeb7cfd1242"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"83Lj+MjGbxWprKMwysOM+N2N5ST2O1jDbpufZz7v/lA2ZL7COQoOjpbjhMlKyyHgf4763UaS/7ey+hR94qrICg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-29T18:23:41.317891Z","bundle_sha256":"948455d1132e39bbb47b71f8d28f516a5227ebaf6b8f6dc1f904a62c73f6a3b5"}}