dataset stringclasses 1
value | example_id stringlengths 33 33 | group_id stringclasses 348
values | input dict | source dict | split stringclasses 1
value | target_descriptors dict | targets dict | split_seed int64 1.23k 1.23k |
|---|---|---|---|---|---|---|---|---|
wmt24_en_ru_rate | wmt24_en_ru_rate:00142e7d1558cc32 | wmt24_en_ru_rate_source:c14f1e441fba173a | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "I hate lava #firetemple",
"system": "refA",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Ненавижу лаву #храмогня"
} | {
"annotation_protocol": "RATE",
"annotator": "b2351f439b1a8fc6d87e9ceffca73643",
"doc_id": "test-en-social_111977498791056432",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 217,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 70,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:005352c3569c4995 | wmt24_en_ru_rate_source:64762b733c64b8a1 | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "Construction waste recycling firm opens new Cumbernauld plant amid circular economy drive",
"system": "Dubformer",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"transl... | {
"annotation_protocol": "RATE",
"annotator": "a88bb75903b6c0f46abc0820ed9a5614",
"doc_id": "test-en-news_scotsman.87448",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 86,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.jsonl",
... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 80,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0103267648017a23 | wmt24_en_ru_rate_source:cbae76ac2cc0c5ea | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "After that a couple of site visits on Saturdays with a client in Swansea.",
"system": "ONLINE-G",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "После ... | {
"annotation_protocol": "RATE",
"annotator": "b2351f439b1a8fc6d87e9ceffca73643",
"doc_id": "test-en-social_112112271633685072",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 409,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 55,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:011152a3e624e574 | wmt24_en_ru_rate_source:ce2f30d0a1e02333 | {
"domain": "speech",
"language_pair": "en-ru",
"reference": null,
"source_text": "The king, Labella, is said to have travelled the 1,600 miles to Jerusalem. Legend has it, when he returned and Jerusalem fell to the Islamic conquest, Lalabella ordered a new home for Christianity. Each church was carved from a s... | {
"annotation_protocol": "RATE",
"annotator": "8fa06b84d697424004852b24f95aec25",
"doc_id": "test-en-speech_JTpvaw6ywbE_000",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 746,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 70,
"score_fluency": 80,
"score_style": 80
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:01401158a2446864 | wmt24_en_ru_rate_source:8b5d1348a5acccc2 | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "The 18 December ministerial summit yielded a carve out for winemakers from the requirement that, by the end of the decade, 10% of products must be supplied in reusable containers within a system for reuse or refill, rising to 40% i... | {
"annotation_protocol": "RATE",
"annotator": "f1ae6a7923cfd75aee476a08631e3785",
"doc_id": "test-en-news_euronews-en.43091",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 31,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.jsonl... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 70,
"score_fluency": 60,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0157a94c777557f7 | wmt24_en_ru_rate_source:9332722463002afa | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "\"THAT'S GONNA BE TRICKY!\"",
"system": "refA",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "«ЭТО НЕ ТАК-ТО ПРОСТО!»"
} | {
"annotation_protocol": "RATE",
"annotator": "411cac0f20b39a086df384631c771cab",
"doc_id": "test-en-literary_fight_above_the_trees_chunk_2_words_991",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 847,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/w... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 95,
"score_fluency": 95,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:015fc5b40206b93f | wmt24_en_ru_rate_source:2865941f3863dc05 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Final edit: that level turned out to be a tool assisted upload, which no one thought was possible at the time but apparently it was! But people still persisted and it got beat in the end! https://fgc.network/objects/0f1b42c6-cbb1... | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_112226149509052560",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 608,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 80,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:018c2d7cf70f576e | wmt24_en_ru_rate_source:65ab1d8ce968fbe5 | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "The workers\" mood is increasingly angry. \"Appeals are circulating with fantasies of revolution,\" warned Mr Habeck. The far-right Alternative for Germany party is doing its best to fan the grievances. In Dresden the Free Saxons, ... | {
"annotation_protocol": "RATE",
"annotator": "411cac0f20b39a086df384631c771cab",
"doc_id": "test-en-news_economist.14223",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 24,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.jsonl",... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 50,
"score_fluency": 40,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:01e73d1ebfa1990d | wmt24_en_ru_rate_source:6a1d83d462b8c129 | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "“Yes sir. There should be on their way now, we gave them time to grab rations from the mess” \tAt 0500 sharp, Exodus company assembled at the motor pool. Shaw stepped onto a box and began to speak.",
"system": "Dubformer",
... | {
"annotation_protocol": "RATE",
"annotator": "411cac0f20b39a086df384631c771cab",
"doc_id": "test-en-literary_the_other_side_stormfall_chunk_2_words_956",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 970,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translatio... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 90,
"score_fluency": 85,
"score_style": 90
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:01fa9145589db79f | wmt24_en_ru_rate_source:026895f1a3738d4f | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Jakob Nielsen’s all-in on AI is melting my brain.",
"system": "refA",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "У меня мозги плавятся от того, как... | {
"annotation_protocol": "RATE",
"annotator": "fb8189bef63a82f96fa1b11a12edc7f5",
"doc_id": "test-en-social_111976217731399552",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 166,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 10,
"score_fluency": 40,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:01fdb6795c05ed83 | wmt24_en_ru_rate_source:01525040bfa1aa52 | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "\"My own flight is very different from being abruptly dragged off by a massive, pink dragon!\"",
"system": "ONLINE-G",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
... | {
"annotation_protocol": "RATE",
"annotator": "8fa06b84d697424004852b24f95aec25",
"doc_id": "test-en-literary_fight_above_the_trees_chunk_1_words_996",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 816,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/w... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0208c333fc620b3e | wmt24_en_ru_rate_source:ed0990f118761e7f | {
"domain": "speech",
"language_pair": "en-ru",
"reference": null,
"source_text": "Proper handling and disposal of waste at your auto repair facility is important for employee health, but also to avoid fines or significant legal settlements. Also, customers want to do business with companies that make an effort... | {
"annotation_protocol": "RATE",
"annotator": "fb8189bef63a82f96fa1b11a12edc7f5",
"doc_id": "test-en-speech_5jn6pLG9P7k_000",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 719,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 90,
"score_fluency": 87,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:02178b4df2038797 | wmt24_en_ru_rate_source:78e1fa27094c72a6 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Don’t travel too far, someone will beat you to it.",
"system": "Unbabel-Tower70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Не путешествуйте слиш... | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-social_112111346044907536",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 381,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 65,
"score_fluency": 95,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:023c4cef4b4767d3 | wmt24_en_ru_rate_source:3ae373b6acbaa31c | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "@user42 this is very interesting because it says my latest public post to my-blog was in 2022 but I am certain this repo has been private for several years",
"system": "Yandex",
"task": "Predict expert multi-dimensional RATE ... | {
"annotation_protocol": "RATE",
"annotator": "f1ae6a7923cfd75aee476a08631e3785",
"doc_id": "test-en-social_112130344809706592",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 475,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 50,
"score_fluency": 90,
"score_style": 90
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:028660907b08529a | wmt24_en_ru_rate_source:af948cd6572af9f1 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "let’s tank the first week of the new year",
"system": "refA",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "обрушим первую неделю нового года"
} | {
"annotation_protocol": "RATE",
"annotator": "a2f0c36c1416bb5993da016e5f7723c9",
"doc_id": "test-en-social_112201850099412432",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 592,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 35,
"score_fluency": 85,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:02893e3205cebb7b | wmt24_en_ru_rate_source:b529a1847a50f0e7 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "I need to document this crap.",
"system": "Yandex",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Мне нужно задокументировать это дерьмо."
} | {
"annotation_protocol": "RATE",
"annotator": "fb8189bef63a82f96fa1b11a12edc7f5",
"doc_id": "test-en-social_111976217731399552",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 168,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 90,
"score_fluency": 30,
"score_style": 80
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:02a42f6118ab3c2a | wmt24_en_ru_rate_source:229eb002bd59c07d | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Being 40-ish, there was the nagging concern that I'd discover kids I didn't know about. So far, just one of my cousins abandoned kids (maybe).",
"system": "Claude-3.5",
"task": "Predict expert multi-dimensional RATE scores fo... | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_112107496062298544",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 248,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:02a8cb1f18a4b137 | wmt24_en_ru_rate_source:2e3c9849735626df | {
"domain": "speech",
"language_pair": "en-ru",
"reference": null,
"source_text": "Knight of the Galactic Railroad by Kenji Miyazawa. This is a Japanese story of two young boys who embark on a mystical journey on a celestial train, going through the galaxy, encountering various allegorical experiences and deep ... | {
"annotation_protocol": "RATE",
"annotator": "8fa06b84d697424004852b24f95aec25",
"doc_id": "test-en-speech_0mytlKOHr74_006",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 694,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 70,
"score_fluency": 65,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:02bd6529383f44af | wmt24_en_ru_rate_source:396dc25ab0487a9c | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "At the precinct level, we just need to look at the Sydney CBD as an example of an area which has grown from its pre-settlement landscape into a two-storey city, from a maximum height of 150 feet to the soaring towers of today. At a... | {
"annotation_protocol": "RATE",
"annotator": "693653dd9a8ddb96a8ef6f6160d82439",
"doc_id": "test-en-news_brisbanetimes.com.au.228963",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 7,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_sp... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 60,
"score_fluency": 30,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:02d0431cebae16a5 | wmt24_en_ru_rate_source:124a3b21f7cd7992 | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "Chapter 1",
"system": "Unbabel-Tower70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Глава 1"
} | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-literary_the_other_side_stormfall_chunk_1_words_992",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 939,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translatio... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:02da4cff0cec6293 | wmt24_en_ru_rate_source:14945b9d35f5074d | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "I've wanted to fly since I was a child.",
"system": "refA",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Я с детства хотела летать."
} | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_112152593528184304",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 495,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:02f352780e11f207 | wmt24_en_ru_rate_source:cdef91f3340fd877 | {
"domain": "speech",
"language_pair": "en-ru",
"reference": null,
"source_text": "Oh my god, you guys, listen, I'm so excited because this comedy special is about 23 years in the making, okay? Yeah, I'll tell you what, what I mean by that. 23 years ago is when I first came to the comedy store and became a regu... | {
"annotation_protocol": "RATE",
"annotator": "411cac0f20b39a086df384631c771cab",
"doc_id": "test-en-speech_OtKhIhSgy78_001",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 761,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 60,
"score_fluency": 60,
"score_style": 85
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0302660e8e3a50a5 | wmt24_en_ru_rate_source:8e89ef6bcfe75d2e | {
"domain": "speech",
"language_pair": "en-ru",
"reference": null,
"source_text": "We're just going to let that sit and rehydrate for a couple of minutes. With ultralight cooking, we're not actually cooking on this dough. We're primarily boiling water so that we can rehydrate our food. It's also one of the reas... | {
"annotation_protocol": "RATE",
"annotator": "8fa06b84d697424004852b24f95aec25",
"doc_id": "test-en-speech_SVWbCPnd0Wc_000",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 777,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 75,
"score_fluency": 55,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:03124f6d2ff57b34 | wmt24_en_ru_rate_source:eca7d453f242b4df | {
"domain": "speech",
"language_pair": "en-ru",
"reference": null,
"source_text": "A long time ago, their parents were killed by poachers and Cheyenne and Kuyu were sold at the market. Taken in by the Help Congo organization, they grew up in semi- captivity with about 30 other orphans out on the islands. One af... | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-speech_2Uo5zKrbips_001",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 702,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 75,
"score_fluency": 95,
"score_style": 90
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0313e72e7812776a | wmt24_en_ru_rate_source:9c6039a43460c921 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Here are a few ways that I am recharged:",
"system": "Claude-3.5",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Вот несколько способов, которыми я во... | {
"annotation_protocol": "RATE",
"annotator": "8fa06b84d697424004852b24f95aec25",
"doc_id": "test-en-social_112107918929771488",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 270,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:03184dd6cc24dc0b | wmt24_en_ru_rate_source:8ad19b68fe1e62e4 | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "A final push for female equality",
"system": "GPT-4",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Финальный рывок к равенству женщин"
} | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-news_csmonitor.com.7750",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 11,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 80,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:032c5e33f8d1e9d2 | wmt24_en_ru_rate_source:2b2aa343faf1e598 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Permission to relax. 📺",
"system": "Yandex",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Разрешите себе расслабиться. 📺"
} | {
"annotation_protocol": "RATE",
"annotator": "77f6d59a810850aeb74146506eb0dbbb",
"doc_id": "test-en-social_112107918929771488",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 275,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:033aa466b68ff0f2 | wmt24_en_ru_rate_source:aa5bf0ae6b79f026 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "lfg $tslq lol",
"system": "Llama3-70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "лфг $тслк лол"
} | {
"annotation_protocol": "RATE",
"annotator": "fb8189bef63a82f96fa1b11a12edc7f5",
"doc_id": "test-en-social_112201850099412432",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 595,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 30,
"score_fluency": 70,
"score_style": 70
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:033db8a76ce6f54c | wmt24_en_ru_rate_source:b7566c6d4a987b5f | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Stumpy and I are home and feeling better now we’ve had a couple of mugs of tea. Back in a fortnight to get the job finished.",
"system": "Claude-3.5",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Ru... | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-social_112140958167999904",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 483,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 90,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0364621011a76f60 | wmt24_en_ru_rate_source:1683ae00e4ba64fc | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "hehehe gotta find moar beer",
"system": "GPT-4",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "хехехе, надо найти ещё пива"
} | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-social_111975617901079872",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 161,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:03697d4c7183537f | wmt24_en_ru_rate_source:8b5d1348a5acccc2 | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "The 18 December ministerial summit yielded a carve out for winemakers from the requirement that, by the end of the decade, 10% of products must be supplied in reusable containers within a system for reuse or refill, rising to 40% i... | {
"annotation_protocol": "RATE",
"annotator": "f1ae6a7923cfd75aee476a08631e3785",
"doc_id": "test-en-news_euronews-en.43091",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 31,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.jsonl... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 65,
"score_fluency": 80,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:036ed0dfdd627693 | wmt24_en_ru_rate_source:365ec463b020368a | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Or not.",
"system": "Claude-3.5",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Или нет."
} | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_112140958167999904",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 484,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:038fd0c41009bbed | wmt24_en_ru_rate_source:1472c4e201bbe1e0 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "\"We're on the surface and transmitting\"",
"system": "GPT-4",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "\"Мы на поверхности и передаем сигнал\""
... | {
"annotation_protocol": "RATE",
"annotator": "b2351f439b1a8fc6d87e9ceffca73643",
"doc_id": "test-en-social_111977766001055104",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 233,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:03a1b200f677c3f7 | wmt24_en_ru_rate_source:e0b3ba1db312699f | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "With the Cost of Living Act legislation ending on 31 March 2024, a new consultation document seeks to extend controls on the level of rent increases that can be levied in the coming year. The consultation ends this Monday and propo... | {
"annotation_protocol": "RATE",
"annotator": "77f6d59a810850aeb74146506eb0dbbb",
"doc_id": "test-en-news_scotsman.87458",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 98,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.jsonl",
... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 85,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:03b9c19c60784deb | wmt24_en_ru_rate_source:e201d987919a735f | {
"domain": "speech",
"language_pair": "en-ru",
"reference": null,
"source_text": "We're not ready. And it's going to be more dangerous out there because you fired Greg. Because we don't have any money to pay him, Burt, you're doing this, not me. hell I am. Do you not understand what I had to do to get these ki... | {
"annotation_protocol": "RATE",
"annotator": "411cac0f20b39a086df384631c771cab",
"doc_id": "test-en-speech_-IPJfZjzCZQ_000",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 685,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 85,
"score_fluency": 90,
"score_style": 95
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:03d07b51cff0b28d | wmt24_en_ru_rate_source:2ae0d0d4db334eb0 | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "Siso's depictions of land, water center new gallery exhibition",
"system": "ONLINE-G",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Выставка \"Изображе... | {
"annotation_protocol": "RATE",
"annotator": "b2351f439b1a8fc6d87e9ceffca73643",
"doc_id": "test-en-news_beverly_press.3585",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 1,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.jsonl... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 10,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:03da34af0206dc0b | wmt24_en_ru_rate_source:4aa7811b64642f56 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "I got new stickers.",
"system": "Unbabel-Tower70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Я получил новые наклейки."
} | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-social_112111346044907536",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 374,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:03fb22bd652a0d8b | wmt24_en_ru_rate_source:b95178ff66ca4a03 | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "While the Louisiana Republican can avoid confronting the issue for now, on January 19 the country will face its deadline to pass four of 12 annual spending bills. Johnson has publicly stated that he would not support a \"continuing... | {
"annotation_protocol": "RATE",
"annotator": "e8f9547a8f759f2ce8e3e4230c33aa06",
"doc_id": "test-en-news_newsweek.63908",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 45,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.jsonl",
... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 30,
"score_fluency": 60,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:03fced7178946aab | wmt24_en_ru_rate_source:b28a6ef00d5004f8 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "I loved cooking and baking for my posse.",
"system": "Claude-3.5",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Я обожал готовить и печь для своей ко... | {
"annotation_protocol": "RATE",
"annotator": "8fa06b84d697424004852b24f95aec25",
"doc_id": "test-en-social_112107889726289648",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 259,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 80,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:040ab45c16a7bd15 | wmt24_en_ru_rate_source:0663c4bea1d35769 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "what why are the GPS coordinates getting rounded to the nearest full degree that's absolutely worthless",
"system": "Claude-3.5",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine transla... | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_112109432154590752",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 295,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 90,
"score_fluency": 80,
"score_style": 80
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:04307fe524ec8a88 | wmt24_en_ru_rate_source:f97d4c946747a461 | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "Moving the surveillance onto my mobile device, I watched each one intently. Nothing. It was like they were never there. That makes no sense. Cameras had tight security, even with remote access. How..? ARGH. I stood angrily, rea... | {
"annotation_protocol": "RATE",
"annotator": "a88bb75903b6c0f46abc0820ed9a5614",
"doc_id": "test-en-literary_detestable_chunk_2_words_945",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 810,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 70,
"score_fluency": 85,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:043271c26f9dad90 | wmt24_en_ru_rate_source:6c796575b0c10495 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "It's my first ever crown dentist visit: Prep today, then fitting in a fortnight. I'm practising opening my mouth wide.",
"system": "Unbabel-Tower70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Ru... | {
"annotation_protocol": "RATE",
"annotator": "f1ae6a7923cfd75aee476a08631e3785",
"doc_id": "test-en-social_112140958167999904",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 481,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 80,
"score_fluency": 75,
"score_style": 95
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:044cc74225e01c39 | wmt24_en_ru_rate_source:3ea2b787e1085269 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "first complaint: we need cute logbooks. black is boring, pink is not even pink, it's salmon.",
"system": "Llama3-70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"... | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_112152593528184304",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 494,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 80,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:046745c532f75ab2 | wmt24_en_ru_rate_source:24fe017ae22abd4d | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "A vocal anarchist who's shit about hiding his identity giving his DNA to a biotechnology company for a dubiouse printout of my \"Ancestry.\" What could go wrong?",
"system": "refA",
"task": "Predict expert multi-dimensional R... | {
"annotation_protocol": "RATE",
"annotator": "f1ae6a7923cfd75aee476a08631e3785",
"doc_id": "test-en-social_112107496062298544",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 242,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 90,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:046b8c3d2e713299 | wmt24_en_ru_rate_source:e6fe42f64b6bf46a | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Not impossible, but it's going to take a VERY long time and require a LOT of extra work.",
"system": "Dubformer",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"trans... | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-social_112152593528184304",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 500,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 80,
"score_style": 90
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0486def414e23cef | wmt24_en_ru_rate_source:03e4d7fb3d256d7c | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "The Ice King looked down at his queen, disgust from his eyes. Maybe, just maybe, he would have mercy on her. The woman who he had been with for twenty-two years, the one who always cared for him at his lowest times, the one who... | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-literary_forever_snow_chunk_1_words_993",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 878,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_r... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 93,
"score_fluency": 75,
"score_style": 80
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:048a165148ba5125 | wmt24_en_ru_rate_source:6800e610fd2cfb5c | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "you know what has semantic class names? css and class names",
"system": "GPT-4",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Знаете, что имеет семан... | {
"annotation_protocol": "RATE",
"annotator": "f6f96602abf0b00db622c8c12294c922",
"doc_id": "test-en-social_112166537145572640",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 573,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 25,
"score_fluency": 35,
"score_style": 50
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:04937c6514338836 | wmt24_en_ru_rate_source:bf54388fa999ce14 | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "“Hey Cohren, does Harris still have Snowball?” Nemic asks as they got settled in.",
"system": "ONLINE-G",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation... | {
"annotation_protocol": "RATE",
"annotator": "f6f96602abf0b00db622c8c12294c922",
"doc_id": "test-en-literary_the_other_side_stormfall_chunk_2_words_956",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 973,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translatio... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 55,
"score_fluency": 40,
"score_style": 60
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:049ea55e966e1f9f | wmt24_en_ru_rate_source:a2518d58f4f4cd7f | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "I was on my period but I was supposed to strip down. So they let me have these mesh underwear you wear after childbirth. But they strip you completely down in the OR and drape you. So at some point a nice nurse replaced my period... | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_111977447547284544",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 186,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 90,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:04ae7bee741a5720 | wmt24_en_ru_rate_source:ef176e113fab1e2c | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "The report also includes a comparison of hull loss accident rates per million departures, which helps account for the fact that some of the models are more common or have been around longer than others. The Airbus A310, which was i... | {
"annotation_protocol": "RATE",
"annotator": "8fa06b84d697424004852b24f95aec25",
"doc_id": "test-en-news_seattle_times.800119",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 137,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 50,
"score_fluency": 80,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:04aeb052bd070ffe | wmt24_en_ru_rate_source:fac0806f0c9c2abb | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "JANUARY 19th, 2:34 PM, 2543.",
"system": "GPT-4",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "19 ЯНВАРЯ, 14:34, 2543 год."
} | {
"annotation_protocol": "RATE",
"annotator": "a88bb75903b6c0f46abc0820ed9a5614",
"doc_id": "test-en-literary_detestable_chunk_2_words_945",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 808,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 80
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:04e51f404a4e6a38 | wmt24_en_ru_rate_source:0dae422f38b3e7d4 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "field 0 is the degree, field 1 is the minute, field 2 is the second",
"system": "refA",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Поле 0 для граду... | {
"annotation_protocol": "RATE",
"annotator": "f6f96602abf0b00db622c8c12294c922",
"doc_id": "test-en-social_112109432154590752",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 298,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 90,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:04f057d23639dddb | wmt24_en_ru_rate_source:397ce86752256b1b | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "A person is here to fix my kitchen floor, so I'm gonna pretend I still have a job and am wfh while editing Too Hot to Handle.",
"system": "Dubformer",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Ru... | {
"annotation_protocol": "RATE",
"annotator": "a88bb75903b6c0f46abc0820ed9a5614",
"doc_id": "test-en-social_112122127346453600",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 452,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 70,
"score_fluency": 60,
"score_style": 97
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:04f86615f1f44475 | wmt24_en_ru_rate_source:b95178ff66ca4a03 | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "While the Louisiana Republican can avoid confronting the issue for now, on January 19 the country will face its deadline to pass four of 12 annual spending bills. Johnson has publicly stated that he would not support a \"continuing... | {
"annotation_protocol": "RATE",
"annotator": "e8f9547a8f759f2ce8e3e4230c33aa06",
"doc_id": "test-en-news_newsweek.63908",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 45,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.jsonl",
... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 75,
"score_fluency": 85,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:05217d9d67faa9d2 | wmt24_en_ru_rate_source:5c18aa9eb7d70f59 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "redid the head!",
"system": "Unbabel-Tower70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "перепроектировал голову!"
} | {
"annotation_protocol": "RATE",
"annotator": "8fa06b84d697424004852b24f95aec25",
"doc_id": "test-en-social_112294019357219888",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 675,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 50,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0547f392153ee170 | wmt24_en_ru_rate_source:365ec463b020368a | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Or not.",
"system": "Dubformer",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Или нет."
} | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_112140958167999904",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 484,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:054a5f0e14549020 | wmt24_en_ru_rate_source:eac5327b53906802 | {
"domain": "speech",
"language_pair": "en-ru",
"reference": null,
"source_text": "603 right now on this Wednesday morning, and we are following a tragic story this morning. A teenager is dead after police found her body in an eastside ditch. The gruesome discovery was made hours after her parents reported her ... | {
"annotation_protocol": "RATE",
"annotator": "f1ae6a7923cfd75aee476a08631e3785",
"doc_id": "test-en-speech_AC2UOuvVol8_000",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 736,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 40,
"score_fluency": 50,
"score_style": 80
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:054b2f8ade69771b | wmt24_en_ru_rate_source:98b38593ca6e38b5 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "https://twitter.com/Ahoyoo_Twitch/status/1771286936944099483",
"system": "Yandex",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "https://twitter.com/A... | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-social_112226149509052560",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 612,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0551e505f5c4ab8b | wmt24_en_ru_rate_source:5382e04b4d3bb66f | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Got some somewhere but exactly where?",
"system": "Claude-3.5",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Где-то есть, но вот где именно?"
} | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-social_112112271633685072",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 414,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 85,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:05619331d9718852 | wmt24_en_ru_rate_source:d6187c2c8b6db210 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "several hours in, every news outlet is headlining the story, and yet the AT&T status site still claims there’s no outage. Slowest static site i’ve ever used…",
"system": "GPT-4",
"task": "Predict expert multi-dimensional RATE... | {
"annotation_protocol": "RATE",
"annotator": "f6f96602abf0b00db622c8c12294c922",
"doc_id": "test-en-social_111976249663909024",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 181,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 50,
"score_fluency": 50,
"score_style": 60
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:05af821ad13357ad | wmt24_en_ru_rate_source:d6187c2c8b6db210 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "several hours in, every news outlet is headlining the story, and yet the AT&T status site still claims there’s no outage. Slowest static site i’ve ever used…",
"system": "Yandex",
"task": "Predict expert multi-dimensional RAT... | {
"annotation_protocol": "RATE",
"annotator": "f6f96602abf0b00db622c8c12294c922",
"doc_id": "test-en-social_111976249663909024",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 181,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 60,
"score_fluency": 60,
"score_style": 70
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:05d80ed3c834f5e8 | wmt24_en_ru_rate_source:e401506c16cb2947 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "@user37 My current collection consists of two green pairs, one blue-purple and one pinkish lilac pair. The last one’s at home currently, but I usually carry at least two pairs with me.",
"system": "Unbabel-Tower70B",
"task": ... | {
"annotation_protocol": "RATE",
"annotator": "f6f96602abf0b00db622c8c12294c922",
"doc_id": "test-en-social_112112980319428992",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 422,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 45,
"score_fluency": 55,
"score_style": 25
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:05dd381f9fc782af | wmt24_en_ru_rate_source:c93475a8eef23777 | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "“All tanks were moving out” The commander started calmly but clearly into his headset. The engines of a platoon of tanks and Strykers roared to life as they started moving forward. The Kronos Cohren was on jolted as its break r... | {
"annotation_protocol": "RATE",
"annotator": "f1ae6a7923cfd75aee476a08631e3785",
"doc_id": "test-en-literary_the_other_side_stormfall_chunk_2_words_956",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 977,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translatio... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 65,
"score_fluency": 70,
"score_style": 80
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:05e34c5361143582 | wmt24_en_ru_rate_source:a6398d2a4b0b4d58 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "“brb just updating the padding on paragraphs boss”",
"system": "Unbabel-Tower70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "\"Скоро вернусь, сэр,... | {
"annotation_protocol": "RATE",
"annotator": "f6f96602abf0b00db622c8c12294c922",
"doc_id": "test-en-social_112166537145572640",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 577,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 75,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0629f631808fe85f | wmt24_en_ru_rate_source:c8e8bfa13ca10572 | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "Politicians get it wrong when it comes to rent caps",
"system": "Llama3-70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Политики ошибаются, когда ре... | {
"annotation_protocol": "RATE",
"annotator": "e8f9547a8f759f2ce8e3e4230c33aa06",
"doc_id": "test-en-news_scotsman.87458",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 94,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.jsonl",
... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 60,
"score_fluency": 85,
"score_style": 85
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:063884e036846248 | wmt24_en_ru_rate_source:f77fc3c4ab659b61 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "First steps are already done.",
"system": "Llama3-70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Первые шаги уже сделаны."
} | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-social_112121157211696272",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 430,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:064aa5d5590d1362 | wmt24_en_ru_rate_source:9332722463002afa | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "\"THAT'S GONNA BE TRICKY!\"",
"system": "Claude-3.5",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "\"ЭТО БУДЕТ НЕПРОСТО!\""
} | {
"annotation_protocol": "RATE",
"annotator": "411cac0f20b39a086df384631c771cab",
"doc_id": "test-en-literary_fight_above_the_trees_chunk_2_words_991",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 847,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/w... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 95,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:06580e61fe053750 | wmt24_en_ru_rate_source:8ecc612df3a8b72b | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "No, clearly not. The creature just hissed and tried to fly directly into Thassalin. Thankfully, despite his size, Thassalin was nimble enough to get out of the way, but Nyssi couldn't help but scream as she very nearly lost her... | {
"annotation_protocol": "RATE",
"annotator": "a88bb75903b6c0f46abc0820ed9a5614",
"doc_id": "test-en-literary_fight_above_the_trees_chunk_2_words_991",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 844,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/w... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 70,
"score_fluency": 60,
"score_style": 70
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:06738ad7df056a9c | wmt24_en_ru_rate_source:4be98c6747e2807d | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "There's another mid-scale version that could've existed between the two that tells the same expanded resonant story but with a tighter runtime and less insane production values. I think that would've been my favorite incarnation ... | {
"annotation_protocol": "RATE",
"annotator": "e8f9547a8f759f2ce8e3e4230c33aa06",
"doc_id": "test-en-social_111975537143453440",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 154,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 70,
"score_fluency": 50,
"score_style": 60
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0688052749d301c5 | wmt24_en_ru_rate_source:a62d4bfdb1e8d6fd | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "In an effort to defuse the tension with the farmers, the government agreed to a gradual removal of the diesel subsidy over three years and to keeping the exemption from the car tax. The farmers pooh-poohed the concessions as insuff... | {
"annotation_protocol": "RATE",
"annotator": "b2351f439b1a8fc6d87e9ceffca73643",
"doc_id": "test-en-news_economist.14223",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 23,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.jsonl",... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 80,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:06d08d6af2a691cb | wmt24_en_ru_rate_source:4369cb9526740ba6 | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "Thus, we had the introduction of the Cost of Living (Tenant protection) (Scotland) Act 2022 nearly 18 months ago which sought to cap rents for a limited period to ease the pressures of the cost-of-living crisis. What we find now fr... | {
"annotation_protocol": "RATE",
"annotator": "77f6d59a810850aeb74146506eb0dbbb",
"doc_id": "test-en-news_scotsman.87458",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 96,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.jsonl",
... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 60,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:06d97a314ef8b35d | wmt24_en_ru_rate_source:9fbba1e1bb57b714 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Maybe a fisheye lens is a bit overkill for a nesting box.",
"system": "Llama3-70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Может быть, объектив... | {
"annotation_protocol": "RATE",
"annotator": "8fa06b84d697424004852b24f95aec25",
"doc_id": "test-en-social_112111193384667328",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 313,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 75,
"score_fluency": 100,
"score_style": 75
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:07269ae6d2c73cfa | wmt24_en_ru_rate_source:4a8a62851e87dd55 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "@user18 what does it check your face against?",
"system": "Claude-3.5",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "@user18 с чем сравнивается ваше ... | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-social_112132847725764752",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 479,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 86,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:072c55406a1128ed | wmt24_en_ru_rate_source:0b97854e252dc1c7 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "this is why we have such strong ties to nostalgia.",
"system": "refA",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Вот почему мы так любим поносталь... | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_111977470885664240",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 194,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:073a66f3bc9302f6 | wmt24_en_ru_rate_source:5acbd01cc79a9981 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "I better get some motorbikes fixed, or else!",
"system": "GPT-4",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Мне лучше починить несколько мотоцикло... | {
"annotation_protocol": "RATE",
"annotator": "f1ae6a7923cfd75aee476a08631e3785",
"doc_id": "test-en-social_112112271633685072",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 410,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 70,
"score_style": 80
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:07693b39d1ba331b | wmt24_en_ru_rate_source:e7b9708c05227cf6 | {
"domain": "speech",
"language_pair": "en-ru",
"reference": null,
"source_text": "Kid Durango of Arizona, the Dust Devils. The layer of dew that covered the prairie, disappeared quickly as the sun rose over the foothills. The moon hung around for a while, and then sunk far below the ranges. The day began like ... | {
"annotation_protocol": "RATE",
"annotator": "fb8189bef63a82f96fa1b11a12edc7f5",
"doc_id": "test-en-speech_MbW7DvLZ9Lg_005",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 753,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 92,
"score_fluency": 80,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:076f821b831f1d63 | wmt24_en_ru_rate_source:9f31360f34f41ee0 | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "\"Kayel!\" Tenuk exclaimed as he glided out of the cloud and began circling, trying to spot Nyssi and Thassalin in the darkness below. \"What do we do? Kayel..? Kayel, where are you?\"",
"system": "GPT-4",
"task": "Predict ... | {
"annotation_protocol": "RATE",
"annotator": "77f6d59a810850aeb74146506eb0dbbb",
"doc_id": "test-en-literary_fight_above_the_trees_chunk_2_words_991",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 851,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/w... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:079cf495f609ff92 | wmt24_en_ru_rate_source:9d41dfd194fa1af2 | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "“Where are moving out to FOB I-131, fifty miles outside of Ianlos. We are to support the ongoing siege and later attack of Ianlos to flush out the insurgency there. From there we will have a foothold and the base of the mountai... | {
"annotation_protocol": "RATE",
"annotator": "a2f0c36c1416bb5993da016e5f7723c9",
"doc_id": "test-en-literary_the_other_side_stormfall_chunk_2_words_956",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 971,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translatio... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 85,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:07a73e631413ecc2 | wmt24_en_ru_rate_source:38ddf091c341b643 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "don't even need a library for it or anything",
"system": "GPT-4",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "даже не нужна для этого библиотека или... | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-social_112166537145572640",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 574,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 70,
"score_style": 90
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:07a7b898b749e50e | wmt24_en_ru_rate_source:c41f2b66da1efea7 | {
"domain": "news",
"language_pair": "en-ru",
"reference": null,
"source_text": "In a Twitter post, political and digital communications expert Akin Akinwale expressed the opinion that the ministry might not have been wrong to use a personal account.",
"system": "Llama3-70B",
"task": "Predict expert multi-d... | {
"annotation_protocol": "RATE",
"annotator": "411cac0f20b39a086df384631c771cab",
"doc_id": "test-en-news_ventures_africa-ar.1773",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 147,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_span... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 70,
"score_fluency": 65,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:07b60d5233044cbf | wmt24_en_ru_rate_source:a241b98f4e44b06b | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Back to this. I hate Queen Gidbo",
"system": "Unbabel-Tower70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Вернемся к этому. Я ненавижу королеву Г... | {
"annotation_protocol": "RATE",
"annotator": "a88bb75903b6c0f46abc0820ed9a5614",
"doc_id": "test-en-social_111977498791056432",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 219,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 60,
"score_fluency": 85,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:07e0e8f7af0bcb18 | wmt24_en_ru_rate_source:6c796575b0c10495 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "It's my first ever crown dentist visit: Prep today, then fitting in a fortnight. I'm practising opening my mouth wide.",
"system": "refA",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machin... | {
"annotation_protocol": "RATE",
"annotator": "f1ae6a7923cfd75aee476a08631e3785",
"doc_id": "test-en-social_112140958167999904",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 481,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 80,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:07f8b4d1cc07516d | wmt24_en_ru_rate_source:7ab6baff79045e93 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "The lander is about the size of old UK telephone box...",
"system": "ONLINE-G",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Посадочный модуль размер... | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-social_111977766001055104",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 227,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 95,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:083aa75612c91307 | wmt24_en_ru_rate_source:a393ffb82aea1a50 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "@user29@user30.engineer oh god, people actually use that outside of trigonometry class???",
"system": "Claude-3.5",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"tra... | {
"annotation_protocol": "RATE",
"annotator": "fb8189bef63a82f96fa1b11a12edc7f5",
"doc_id": "test-en-social_112109432154590752",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 300,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 70,
"score_style": 60
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:084b794a6f43f686 | wmt24_en_ru_rate_source:124a3b21f7cd7992 | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "Chapter 1",
"system": "Yandex",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Глава 1"
} | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-literary_the_other_side_stormfall_chunk_1_words_992",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 939,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translatio... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:085f40b1695353f1 | wmt24_en_ru_rate_source:99bef207e83b1ef5 | {
"domain": "speech",
"language_pair": "en-ru",
"reference": null,
"source_text": "It is December 1997, and the Imperial Sugar Company is acquiring a new production site at Port Wentworth, from Savannah Foods and Industries Incorporated. There is nothing really of note here, it was doing what businesses do, and... | {
"annotation_protocol": "RATE",
"annotator": "e8f9547a8f759f2ce8e3e4230c33aa06",
"doc_id": "test-en-speech_RyVaAQ9vtkw_000",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 775,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 80,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:086f15b090d8dbba | wmt24_en_ru_rate_source:aa5bf0ae6b79f026 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "lfg $tslq lol",
"system": "ONLINE-G",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "lfg $tslq лол"
} | {
"annotation_protocol": "RATE",
"annotator": "fb8189bef63a82f96fa1b11a12edc7f5",
"doc_id": "test-en-social_112201850099412432",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 595,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 66,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:08850a40d6058fed | wmt24_en_ru_rate_source:b76aae4dd389fc91 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "i do not like this",
"system": "Yandex",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "мне это не нравится"
} | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_112109432154590752",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 299,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:088597298ca441ba | wmt24_en_ru_rate_source:a2f16a115002ced6 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "📉",
"system": "ONLINE-G",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "📉"
} | {
"annotation_protocol": "RATE",
"annotator": "a2f0c36c1416bb5993da016e5f7723c9",
"doc_id": "test-en-social_112201850099412432",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 593,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:089645972e7996b9 | wmt24_en_ru_rate_source:c55f3f62450489c1 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Does this make sense? Would this be useful?",
"system": "refA",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Логичный вариант? Будет он кому-нибудь п... | {
"annotation_protocol": "RATE",
"annotator": "fb8189bef63a82f96fa1b11a12edc7f5",
"doc_id": "test-en-social_112289379466442912",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 654,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 50,
"score_fluency": 90,
"score_style": 90
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:089fde8433882520 | wmt24_en_ru_rate_source:58fbf0362f12f762 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Very niche issue but it kinda sucks these days that if you're interested in building/customising lil robot figures you HAVE to build something made for war and killing things",
"system": "refA",
"task": "Predict expert multi-... | {
"annotation_protocol": "RATE",
"annotator": "b2351f439b1a8fc6d87e9ceffca73643",
"doc_id": "test-en-social_112106953594747568",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 234,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 60,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:08c51289dbad4301 | wmt24_en_ru_rate_source:5381134d68f73928 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Training muscles, not just the mind. 💪",
"system": "Dubformer",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Тренируются мышцы, а не только разум. �... | {
"annotation_protocol": "RATE",
"annotator": "b2351f439b1a8fc6d87e9ceffca73643",
"doc_id": "test-en-social_112107918929771488",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 273,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 70,
"score_fluency": 70,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:08d1c8e827a505dd | wmt24_en_ru_rate_source:64572319a8e46de1 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "Might be the odd cheater in my family, but considering my largely broke folks' roots, any affairs were probably in mostly homogenous villages, small towns, and ethnic ghettos lol.",
"system": "ONLINE-G",
"task": "Predict expe... | {
"annotation_protocol": "RATE",
"annotator": "f1ae6a7923cfd75aee476a08631e3785",
"doc_id": "test-en-social_112107496062298544",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 244,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 50,
"score_fluency": 90,
"score_style": 75
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:08e207c01ba9c349 | wmt24_en_ru_rate_source:cdea0b7e7b589556 | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "If I can get the file to fucking open. (Fuck One Drive.)",
"system": "GPT-4",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Если я смогу заставить это... | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_112122127346453600",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 453,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:08eb4128df52fca4 | wmt24_en_ru_rate_source:996fabcc4664fd0b | {
"domain": "speech",
"language_pair": "en-ru",
"reference": null,
"source_text": "And he said, \" You have a tattoo.\" And I said, I have a few on your hand, I have a banjo, and I told him about it, as a friend of mine, Darryl Adams said he said I had to. Oh yeah, there's a banjo right there on your hand. Can ... | {
"annotation_protocol": "RATE",
"annotator": "a2f0c36c1416bb5993da016e5f7723c9",
"doc_id": "test-en-speech_6D9Z7antjMw_002",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 724,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.json... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 40,
"score_fluency": 90,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:093966df261b7c44 | wmt24_en_ru_rate_source:54624aeabdc3f67a | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "So, my life kind of sucks right now.",
"system": "refA",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Ну в общем у меня сейчас все плохо."
} | {
"annotation_protocol": "RATE",
"annotator": "411cac0f20b39a086df384631c771cab",
"doc_id": "test-en-social_112109186993725712",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 279,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 90,
"score_style": 90
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0948669f3fac3734 | wmt24_en_ru_rate_source:14945b9d35f5074d | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "I've wanted to fly since I was a child.",
"system": "Unbabel-Tower70B",
"task": "Predict expert multi-dimensional RATE scores for an English-to-Russian machine translation.",
"translation": "Я мечтал летать с самого детства... | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_112152593528184304",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 495,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 100,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0955c7a112c2c480 | wmt24_en_ru_rate_source:9577cfac41d2f48f | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "That is to say, it is a delight, she employed all of this excellent skill in a service of a larger social point this week.",
"system": "Unbabel-Tower70B",
"task": "Predict expert multi-dimensional RATE scores for an English-t... | {
"annotation_protocol": "RATE",
"annotator": "7cd5e9f4a9a624a87947d236cc8851d9",
"doc_id": "test-en-social_112110250600185424",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 305,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 70,
"score_fluency": 100,
"score_style": 100
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:0976c7a9b7651f54 | wmt24_en_ru_rate_source:1744189e03378660 | {
"domain": "literary",
"language_pair": "en-ru",
"reference": null,
"source_text": "Thassalin hissed, then circled round, closer to the mountain he called home. The dark beast followed, but somewhat panicked as a blast of fire hit it from behind. The creature hadn't realised it was no longer alone. It panicked... | {
"annotation_protocol": "RATE",
"annotator": "ef6d53a3cb6f0e8ab7719d77626b2f81",
"doc_id": "test-en-literary_fight_above_the_trees_chunk_2_words_991",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 848,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/w... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 67,
"score_fluency": 50,
"score_style": 70
} | 1,234 |
wmt24_en_ru_rate | wmt24_en_ru_rate:099e19c246c836ba | wmt24_en_ru_rate_source:2a838b2077d544ff | {
"domain": "social",
"language_pair": "en-ru",
"reference": null,
"source_text": "with no reason to question the pedestal we put them on, those parasocial relationships became prominent building blocks that shaped our lives.",
"system": "Unbabel-Tower70B",
"task": "Predict expert multi-dimensional RATE sco... | {
"annotation_protocol": "RATE",
"annotator": "f1ae6a7923cfd75aee476a08631e3785",
"doc_id": "test-en-social_111977470885664240",
"hf_dataset": "yandex/wmt24-en-ru-rate",
"line_id": 193,
"paper": "Refined Assessment for Translation Evaluation",
"raw_file": "data/mvp/raw/translation/wmt24_en_ru_rate_spans.j... | train | {
"score_accuracy": "Expert RATE score for meaning preservation and source-target adequacy on a 0-100 scale.",
"score_fluency": "Expert RATE score for target-language grammaticality, readability, and naturalness on a 0-100 scale.",
"score_style": "Expert RATE score for preserving the source style in the translati... | {
"score_accuracy": 80,
"score_fluency": 60,
"score_style": 80
} | 1,234 |
End of preview. Expand in Data Studio
WMT24 English-Russian RATE
Task-grouped multidimensional machine-translation quality data from RATE.
Contents
The release contains 3,975 complete artifact rows from 497 source-segment groups.
One curated row was removed from the 3,976-row release because its translation was blank and its three scores were zero. The targets are score_accuracy, score_fluency, and score_style, each on a 0--100 scale.
Split organization
Each seed has task-grouped train, validation, and test splits. All candidate translations for one group_id source segment remain in exactly one split, so the same source text never crosses a split boundary. There is no OOD split because RATE does not define a clean unseen-system or unseen-domain collection.
| Seed | Train rows | Validation rows | Test rows |
|---|---|---|---|
| 42 | 2,783 | 400 | 792 |
| 1234 | 2,783 | 400 | 792 |
| 2026 | 2,784 | 399 | 792 |
The source-segment group counts are 348 train, 50 validation, and 99 test groups for every seed.
Provenance
Paper: https://aclanthology.org/2025.findings-emnlp.1203/ Original data: https://huggingface.co/datasets/yandex/wmt24-en-ru-rate
Rebuild with:
python data/rate_quality/build_huggingface_dataset.py \
--source-file data/mvp/curated/wmt24_en_ru_rate.jsonl
- Downloads last month
- 25