{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:5USESHWS7MQGYSZTI6QPNNAE7R","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":"0920834939ff9f5e29c0bb0705dd61e5b7a1bb807294816e51d2f79713687f3e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T17:54:57Z","title_canon_sha256":"8e940bbb35048b9f7ca42b31dd901e0f956426d0277ed57b7279c82ce111e0d1"},"schema_version":"1.0","source":{"id":"2305.14324","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.14324","created_at":"2026-07-05T07:01:41Z"},{"alias_kind":"arxiv_version","alias_value":"2305.14324v2","created_at":"2026-07-05T07:01:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14324","created_at":"2026-07-05T07:01:41Z"},{"alias_kind":"pith_short_12","alias_value":"5USESHWS7MQG","created_at":"2026-07-05T07:01:41Z"},{"alias_kind":"pith_short_16","alias_value":"5USESHWS7MQGYSZT","created_at":"2026-07-05T07:01:41Z"},{"alias_kind":"pith_short_8","alias_value":"5USESHWS","created_at":"2026-07-05T07:01:41Z"}],"graph_snapshots":[{"event_id":"sha256:814891d28f8c2b18478b3bc4d113d07f1e77e9fbc02e1b955d0a4c7175ed978a","target":"graph","created_at":"2026-07-05T07:01:41Z","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/2305.14324/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Kendall's tau is frequently used to meta-evaluate how well machine translation (MT) evaluation metrics score individual translations. Its focus on pairwise score comparisons is intuitive but raises the question of how ties should be handled, a gray area that has motivated different variants in the literature. We demonstrate that, in settings like modern MT meta-evaluation, existing variants have weaknesses arising from their handling of ties, and in some situations can even be gamed. We propose instead to meta-evaluate metrics with a version of pairwise accuracy that gives metrics credit for c","authors_text":"Daniel Deutsch, George Foster, Markus Freitag","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T17:54:57Z","title":"Ties Matter: Meta-Evaluating Modern Metrics with Pairwise Accuracy and Tie Calibration"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14324","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":"sha256:548415876cd9fccb6f92ef7d0e6405dbe15368085b6d308a12dcd2af4586e98f","target":"record","created_at":"2026-07-05T07:01:41Z","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":"0920834939ff9f5e29c0bb0705dd61e5b7a1bb807294816e51d2f79713687f3e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T17:54:57Z","title_canon_sha256":"8e940bbb35048b9f7ca42b31dd901e0f956426d0277ed57b7279c82ce111e0d1"},"schema_version":"1.0","source":{"id":"2305.14324","kind":"arxiv","version":2}},"canonical_sha256":"ed24491ed2fb206c4b3347a0f6b404fc5c26af5e846e84687c91ccfb881ee4ce","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ed24491ed2fb206c4b3347a0f6b404fc5c26af5e846e84687c91ccfb881ee4ce","first_computed_at":"2026-07-05T07:01:41.632906Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:01:41.632906Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"83K9qgm7HUpjTirEjlCNh1EYKfs6TDrJwgOt4PnTiymeG9x8ummVE8EGpXV8GsM2pch+gkYYUx+0PMTSvjiUDw==","signature_status":"signed_v1","signed_at":"2026-07-05T07:01:41.633325Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.14324","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:548415876cd9fccb6f92ef7d0e6405dbe15368085b6d308a12dcd2af4586e98f","sha256:814891d28f8c2b18478b3bc4d113d07f1e77e9fbc02e1b955d0a4c7175ed978a"],"state_sha256":"042a1f9b78f0f314aa51c2b0a3a840d13ce107b4181acbec5464fe332f82c6b1"}