{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ZZMS3DVZTVX64EAP2FZ47ORJP7","short_pith_number":"pith:ZZMS3DVZ","canonical_record":{"source":{"id":"2507.12724","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-17T02:02:54Z","cross_cats_sorted":[],"title_canon_sha256":"c18418f3ce144a5a7501ea986ef044422b04b9e771339e3ad03f14e2592b021d","abstract_canon_sha256":"b39e9526bca6b3f819091f9c573d445ad4d18415c3434f34467d3eaf5a02b4fe"},"schema_version":"1.0"},"canonical_sha256":"ce592d8eb99d6fee100fd173cfba297ff7ae5f6ff66c0b72ef21cadc13df46fd","source":{"kind":"arxiv","id":"2507.12724","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.12724","created_at":"2026-07-05T11:38:37Z"},{"alias_kind":"arxiv_version","alias_value":"2507.12724v1","created_at":"2026-07-05T11:38:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.12724","created_at":"2026-07-05T11:38:37Z"},{"alias_kind":"pith_short_12","alias_value":"ZZMS3DVZTVX6","created_at":"2026-07-05T11:38:37Z"},{"alias_kind":"pith_short_16","alias_value":"ZZMS3DVZTVX64EAP","created_at":"2026-07-05T11:38:37Z"},{"alias_kind":"pith_short_8","alias_value":"ZZMS3DVZ","created_at":"2026-07-05T11:38:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ZZMS3DVZTVX64EAP2FZ47ORJP7","target":"record","payload":{"canonical_record":{"source":{"id":"2507.12724","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-17T02:02:54Z","cross_cats_sorted":[],"title_canon_sha256":"c18418f3ce144a5a7501ea986ef044422b04b9e771339e3ad03f14e2592b021d","abstract_canon_sha256":"b39e9526bca6b3f819091f9c573d445ad4d18415c3434f34467d3eaf5a02b4fe"},"schema_version":"1.0"},"canonical_sha256":"ce592d8eb99d6fee100fd173cfba297ff7ae5f6ff66c0b72ef21cadc13df46fd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:37.911059Z","signature_b64":"HhUUdE0P/Y/UnAcX+T1Fj4T6KPUgLd80zHv6CxWfaJT3caGrrGh7IrcMuxF0lPQTdxc9QwzjuRPtbucqklXIDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce592d8eb99d6fee100fd173cfba297ff7ae5f6ff66c0b72ef21cadc13df46fd","last_reissued_at":"2026-07-05T11:38:37.910539Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:37.910539Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.12724","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":"2026-07-05T11:38:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2DNX8/20CUmsh9UyybTTw3R0Wgu/mjAHQ0cmvrOCCdc9XZ/onSB7InsBjg7KMiKnHzuFOdyzJxXFHGZzonNvBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T00:06:39.887935Z"},"content_sha256":"dbbada7b99e3aee3f946683390e62d0a77f11ea4d97aefba443945a759cc91ae","schema_version":"1.0","event_id":"sha256:dbbada7b99e3aee3f946683390e62d0a77f11ea4d97aefba443945a759cc91ae"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ZZMS3DVZTVX64EAP2FZ47ORJP7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TransEvalnia: Reasoning-based Evaluation and Ranking of Translations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Llion Jones, Richard Sproat, Tianyu Zhao","submitted_at":"2025-07-17T02:02:54Z","abstract_excerpt":"We present TransEvalnia, a prompting-based translation evaluation and ranking system that uses reasoning in performing its evaluations and ranking. This system presents fine-grained evaluations based on a subset of the Multidimensional Quality Metrics (https://themqm.org/), returns an assessment of which translation it deems the best, and provides numerical scores for the various dimensions and for the overall translation. We show that TransEvalnia performs as well as or better than the state-of-the-art MT-Ranker (Moosa et al. 2024) on our own English-Japanese data as well as several language "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.12724","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/2507.12724/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":"2026-07-05T11:38:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y+JOuetr3w6GSwaJu0mUrcTbBmFNC+kvrptAv5uCPykBYiuUmQjii8AlakmkF4bK0OyOux6TrAHxNzAcD9kHCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T00:06:39.888891Z"},"content_sha256":"9226ef790385a491d61efdcbc9b4d2b63adefbfbd8a612df02b68cf78a08609d","schema_version":"1.0","event_id":"sha256:9226ef790385a491d61efdcbc9b4d2b63adefbfbd8a612df02b68cf78a08609d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZZMS3DVZTVX64EAP2FZ47ORJP7/bundle.json","state_url":"https://pith.science/pith/ZZMS3DVZTVX64EAP2FZ47ORJP7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZZMS3DVZTVX64EAP2FZ47ORJP7/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":"2026-08-09T00:06:39Z","links":{"resolver":"https://pith.science/pith/ZZMS3DVZTVX64EAP2FZ47ORJP7","bundle":"https://pith.science/pith/ZZMS3DVZTVX64EAP2FZ47ORJP7/bundle.json","state":"https://pith.science/pith/ZZMS3DVZTVX64EAP2FZ47ORJP7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZZMS3DVZTVX64EAP2FZ47ORJP7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ZZMS3DVZTVX64EAP2FZ47ORJP7","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":"b39e9526bca6b3f819091f9c573d445ad4d18415c3434f34467d3eaf5a02b4fe","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-17T02:02:54Z","title_canon_sha256":"c18418f3ce144a5a7501ea986ef044422b04b9e771339e3ad03f14e2592b021d"},"schema_version":"1.0","source":{"id":"2507.12724","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.12724","created_at":"2026-07-05T11:38:37Z"},{"alias_kind":"arxiv_version","alias_value":"2507.12724v1","created_at":"2026-07-05T11:38:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.12724","created_at":"2026-07-05T11:38:37Z"},{"alias_kind":"pith_short_12","alias_value":"ZZMS3DVZTVX6","created_at":"2026-07-05T11:38:37Z"},{"alias_kind":"pith_short_16","alias_value":"ZZMS3DVZTVX64EAP","created_at":"2026-07-05T11:38:37Z"},{"alias_kind":"pith_short_8","alias_value":"ZZMS3DVZ","created_at":"2026-07-05T11:38:37Z"}],"graph_snapshots":[{"event_id":"sha256:9226ef790385a491d61efdcbc9b4d2b63adefbfbd8a612df02b68cf78a08609d","target":"graph","created_at":"2026-07-05T11:38:37Z","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/2507.12724/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present TransEvalnia, a prompting-based translation evaluation and ranking system that uses reasoning in performing its evaluations and ranking. This system presents fine-grained evaluations based on a subset of the Multidimensional Quality Metrics (https://themqm.org/), returns an assessment of which translation it deems the best, and provides numerical scores for the various dimensions and for the overall translation. We show that TransEvalnia performs as well as or better than the state-of-the-art MT-Ranker (Moosa et al. 2024) on our own English-Japanese data as well as several language ","authors_text":"Llion Jones, Richard Sproat, Tianyu Zhao","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-17T02:02:54Z","title":"TransEvalnia: Reasoning-based Evaluation and Ranking of Translations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.12724","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:dbbada7b99e3aee3f946683390e62d0a77f11ea4d97aefba443945a759cc91ae","target":"record","created_at":"2026-07-05T11:38:37Z","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":"b39e9526bca6b3f819091f9c573d445ad4d18415c3434f34467d3eaf5a02b4fe","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-17T02:02:54Z","title_canon_sha256":"c18418f3ce144a5a7501ea986ef044422b04b9e771339e3ad03f14e2592b021d"},"schema_version":"1.0","source":{"id":"2507.12724","kind":"arxiv","version":1}},"canonical_sha256":"ce592d8eb99d6fee100fd173cfba297ff7ae5f6ff66c0b72ef21cadc13df46fd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ce592d8eb99d6fee100fd173cfba297ff7ae5f6ff66c0b72ef21cadc13df46fd","first_computed_at":"2026-07-05T11:38:37.910539Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:38:37.910539Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HhUUdE0P/Y/UnAcX+T1Fj4T6KPUgLd80zHv6CxWfaJT3caGrrGh7IrcMuxF0lPQTdxc9QwzjuRPtbucqklXIDg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:38:37.911059Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.12724","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dbbada7b99e3aee3f946683390e62d0a77f11ea4d97aefba443945a759cc91ae","sha256:9226ef790385a491d61efdcbc9b4d2b63adefbfbd8a612df02b68cf78a08609d"],"state_sha256":"58e1f79eb2bd50cd3dfa8c924b03afc2a8a0238fd3b7f881a8842ec4c73a8dd1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ZRRJl3r+fOJ4UzcvgxnKLRZU3uY472a5oafO31KARFL1+7tx86gm25OFvWMgw4PSuC5nOiWgpe7FNNrSJnhmBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T00:06:39.895283Z","bundle_sha256":"16ef30581d71a02de47fe6dbc26876865b4457527a50223bc5224fa77be79d16"}}