{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:VEH7SXLZYATEIGZNJW7TWWSXGT","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":"10d7399cbdd00b75ce0644a73efcfe21a451815cca4ebeb14eb8cf7d4c2f43ec","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-01T20:50:13Z","title_canon_sha256":"bc44b9cc29a19e6925b8a8d61ba35c367c75a2891adc71982a2393aa216d41c0"},"schema_version":"1.0","source":{"id":"2404.01474","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.01474","created_at":"2026-07-05T08:03:14Z"},{"alias_kind":"arxiv_version","alias_value":"2404.01474v1","created_at":"2026-07-05T08:03:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.01474","created_at":"2026-07-05T08:03:14Z"},{"alias_kind":"pith_short_12","alias_value":"VEH7SXLZYATE","created_at":"2026-07-05T08:03:14Z"},{"alias_kind":"pith_short_16","alias_value":"VEH7SXLZYATEIGZN","created_at":"2026-07-05T08:03:14Z"},{"alias_kind":"pith_short_8","alias_value":"VEH7SXLZ","created_at":"2026-07-05T08:03:14Z"}],"graph_snapshots":[{"event_id":"sha256:2f478619cdf5555de5a988d2e3b18a92375ef696d160bab2b74d47d600491c43","target":"graph","created_at":"2026-07-05T08:03:14Z","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/2404.01474/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reliable human evaluation is critical to the development of successful natural language generation models, but achieving it is notoriously difficult. Stability is a crucial requirement when ranking systems by quality: consistent ranking of systems across repeated evaluations is not just desirable, but essential. Without it, there is no reliable foundation for hill-climbing or product launch decisions. In this paper, we use machine translation and its state-of-the-art human evaluation framework, MQM, as a case study to understand how to set up reliable human evaluations that yield stable conclu","authors_text":"Ali Dabirmoghaddam, Daniel Deutsch, George Foster, Markus Freitag, Parker Riley, Viresh Ratnakar","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-01T20:50:13Z","title":"Finding Replicable Human Evaluations via Stable Ranking Probability"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.01474","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":"sha256:61ee3cea6056006aa08ed0fecafe49a65a2a66a3eaed67696ae50b6bf0ab3a64","target":"record","created_at":"2026-07-05T08:03:14Z","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":"10d7399cbdd00b75ce0644a73efcfe21a451815cca4ebeb14eb8cf7d4c2f43ec","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-01T20:50:13Z","title_canon_sha256":"bc44b9cc29a19e6925b8a8d61ba35c367c75a2891adc71982a2393aa216d41c0"},"schema_version":"1.0","source":{"id":"2404.01474","kind":"arxiv","version":1}},"canonical_sha256":"a90ff95d79c026441b2d4dbf3b5a5734eb69520bfff876f0795ae3fe7e145fb4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a90ff95d79c026441b2d4dbf3b5a5734eb69520bfff876f0795ae3fe7e145fb4","first_computed_at":"2026-07-05T08:03:14.711857Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:03:14.711857Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YpykJXja/srsE+Y5ADjRHfYnATmrru9r1w3YgfnmuIXL7UiqpMryQeS3u1gmP0EDg6o5TFrctE0dwc2AM0U+CA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:03:14.712349Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.01474","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:61ee3cea6056006aa08ed0fecafe49a65a2a66a3eaed67696ae50b6bf0ab3a64","sha256:2f478619cdf5555de5a988d2e3b18a92375ef696d160bab2b74d47d600491c43"],"state_sha256":"4af8b113d50b81ae7e518d8b6915f9bb3f8d0816ad925180afc887149804119e"}