{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:HSBNYZIZU42XF7LPT4V6HOTFLP","short_pith_number":"pith:HSBNYZIZ","canonical_record":{"source":{"id":"2506.19571","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-24T12:35:00Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1a74a5cda9c420d2c423a28b98460ebfd3e13d849973060d706fa21285503bb8","abstract_canon_sha256":"b3cbd74c36c6d956b75f6681c74d04dacd97cd5a9f9f2bb98e091be8df9cdae4"},"schema_version":"1.0"},"canonical_sha256":"3c82dc6519a73572fd6f9f2be3ba655be68589ba9ebbacd8e57934ccb40310d0","source":{"kind":"arxiv","id":"2506.19571","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.19571","created_at":"2026-07-05T11:26:34Z"},{"alias_kind":"arxiv_version","alias_value":"2506.19571v1","created_at":"2026-07-05T11:26:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.19571","created_at":"2026-07-05T11:26:34Z"},{"alias_kind":"pith_short_12","alias_value":"HSBNYZIZU42X","created_at":"2026-07-05T11:26:34Z"},{"alias_kind":"pith_short_16","alias_value":"HSBNYZIZU42XF7LP","created_at":"2026-07-05T11:26:34Z"},{"alias_kind":"pith_short_8","alias_value":"HSBNYZIZ","created_at":"2026-07-05T11:26:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:HSBNYZIZU42XF7LPT4V6HOTFLP","target":"record","payload":{"canonical_record":{"source":{"id":"2506.19571","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-24T12:35:00Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1a74a5cda9c420d2c423a28b98460ebfd3e13d849973060d706fa21285503bb8","abstract_canon_sha256":"b3cbd74c36c6d956b75f6681c74d04dacd97cd5a9f9f2bb98e091be8df9cdae4"},"schema_version":"1.0"},"canonical_sha256":"3c82dc6519a73572fd6f9f2be3ba655be68589ba9ebbacd8e57934ccb40310d0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:26:34.786260Z","signature_b64":"pEh8rRschs9+T0pFfALkB2BNheXQbk7N+ZcCuv9ROwi6qpuxO/6mfrumOtG0BR7MZ82430g8Mmwmmw5pfj9uAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c82dc6519a73572fd6f9f2be3ba655be68589ba9ebbacd8e57934ccb40310d0","last_reissued_at":"2026-07-05T11:26:34.785785Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:26:34.785785Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.19571","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":"2026-07-05T11:26:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kehZp2TFEkApUDpn9TF46JpPIazkEi9KFabgrdbGKfVfYmGH/2Bde3tL21Sv27p6RSdvtqIjNYJgZX5Xz9nyBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T09:51:10.762216Z"},"content_sha256":"3ace1dd6f2299be3891c1e897bdfcb92230c2135a4e7f9e0931317c835b7a1a6","schema_version":"1.0","event_id":"sha256:3ace1dd6f2299be3891c1e897bdfcb92230c2135a4e7f9e0931317c835b7a1a6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:HSBNYZIZU42XF7LPT4V6HOTFLP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Has Machine Translation Evaluation Achieved Human Parity? The Human Reference and the Limits of Progress","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Lorenzo Proietti, Roberto Navigli, Stefano Perrella","submitted_at":"2025-06-24T12:35:00Z","abstract_excerpt":"In Machine Translation (MT) evaluation, metric performance is assessed based on agreement with human judgments. In recent years, automatic metrics have demonstrated increasingly high levels of agreement with humans. To gain a clearer understanding of metric performance and establish an upper bound, we incorporate human baselines in the MT meta-evaluation, that is, the assessment of MT metrics' capabilities. Our results show that human annotators are not consistently superior to automatic metrics, with state-of-the-art metrics often ranking on par with or higher than human baselines. Despite th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.19571","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/2506.19571/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":"2026-07-05T11:26:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L9s36ZSDx7M1kYooDtxGG7ZjdXyDZamr7TS6VkFiln1jh7k3r3OXnZPadC7c2BgTN/PkCaDJeOIw/gsTWwYQBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T09:51:10.762800Z"},"content_sha256":"8a89d37adf04772833c8633ee710fb20f53bcf035a14f093fdb7f5d937a97eb1","schema_version":"1.0","event_id":"sha256:8a89d37adf04772833c8633ee710fb20f53bcf035a14f093fdb7f5d937a97eb1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HSBNYZIZU42XF7LPT4V6HOTFLP/bundle.json","state_url":"https://pith.science/pith/HSBNYZIZU42XF7LPT4V6HOTFLP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HSBNYZIZU42XF7LPT4V6HOTFLP/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":"2026-08-13T09:51:10Z","links":{"resolver":"https://pith.science/pith/HSBNYZIZU42XF7LPT4V6HOTFLP","bundle":"https://pith.science/pith/HSBNYZIZU42XF7LPT4V6HOTFLP/bundle.json","state":"https://pith.science/pith/HSBNYZIZU42XF7LPT4V6HOTFLP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HSBNYZIZU42XF7LPT4V6HOTFLP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:HSBNYZIZU42XF7LPT4V6HOTFLP","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":"b3cbd74c36c6d956b75f6681c74d04dacd97cd5a9f9f2bb98e091be8df9cdae4","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-24T12:35:00Z","title_canon_sha256":"1a74a5cda9c420d2c423a28b98460ebfd3e13d849973060d706fa21285503bb8"},"schema_version":"1.0","source":{"id":"2506.19571","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.19571","created_at":"2026-07-05T11:26:34Z"},{"alias_kind":"arxiv_version","alias_value":"2506.19571v1","created_at":"2026-07-05T11:26:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.19571","created_at":"2026-07-05T11:26:34Z"},{"alias_kind":"pith_short_12","alias_value":"HSBNYZIZU42X","created_at":"2026-07-05T11:26:34Z"},{"alias_kind":"pith_short_16","alias_value":"HSBNYZIZU42XF7LP","created_at":"2026-07-05T11:26:34Z"},{"alias_kind":"pith_short_8","alias_value":"HSBNYZIZ","created_at":"2026-07-05T11:26:34Z"}],"graph_snapshots":[{"event_id":"sha256:8a89d37adf04772833c8633ee710fb20f53bcf035a14f093fdb7f5d937a97eb1","target":"graph","created_at":"2026-07-05T11:26:34Z","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/2506.19571/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In Machine Translation (MT) evaluation, metric performance is assessed based on agreement with human judgments. In recent years, automatic metrics have demonstrated increasingly high levels of agreement with humans. To gain a clearer understanding of metric performance and establish an upper bound, we incorporate human baselines in the MT meta-evaluation, that is, the assessment of MT metrics' capabilities. Our results show that human annotators are not consistently superior to automatic metrics, with state-of-the-art metrics often ranking on par with or higher than human baselines. Despite th","authors_text":"Lorenzo Proietti, Roberto Navigli, Stefano Perrella","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-24T12:35:00Z","title":"Has Machine Translation Evaluation Achieved Human Parity? The Human Reference and the Limits of Progress"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.19571","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":"sha256:3ace1dd6f2299be3891c1e897bdfcb92230c2135a4e7f9e0931317c835b7a1a6","target":"record","created_at":"2026-07-05T11:26:34Z","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":"b3cbd74c36c6d956b75f6681c74d04dacd97cd5a9f9f2bb98e091be8df9cdae4","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-24T12:35:00Z","title_canon_sha256":"1a74a5cda9c420d2c423a28b98460ebfd3e13d849973060d706fa21285503bb8"},"schema_version":"1.0","source":{"id":"2506.19571","kind":"arxiv","version":1}},"canonical_sha256":"3c82dc6519a73572fd6f9f2be3ba655be68589ba9ebbacd8e57934ccb40310d0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3c82dc6519a73572fd6f9f2be3ba655be68589ba9ebbacd8e57934ccb40310d0","first_computed_at":"2026-07-05T11:26:34.785785Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:26:34.785785Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pEh8rRschs9+T0pFfALkB2BNheXQbk7N+ZcCuv9ROwi6qpuxO/6mfrumOtG0BR7MZ82430g8Mmwmmw5pfj9uAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:26:34.786260Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.19571","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3ace1dd6f2299be3891c1e897bdfcb92230c2135a4e7f9e0931317c835b7a1a6","sha256:8a89d37adf04772833c8633ee710fb20f53bcf035a14f093fdb7f5d937a97eb1"],"state_sha256":"e178414756fbf02df5db491148c11293028f16aca694efba1613676d3d4d3623"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kyefCj+xiJx9ScXPZTmX7tvYAgKEa5e6T64/KigpZmmH3E1hodwrrbYka02eL3a8iCeQXbLia1lW252w5Vi6DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T09:51:10.768397Z","bundle_sha256":"6ac6e31dc06c8f4265f21dcd73a7ce4816ab4014c50691ece116084fccf903ff"}}