Benchmark Almanac

CLEF JOKER 2024 Task 3: Pun Translation

Arampatzis_GoogleTranslate 91.8 tops its leaderboard
Not in model cards
Version 2024

Models translate English puns into French. Organizers compare their output with 832 human translations of 376 source texts using BLEU and BERTScore.

Ten teams submitted 20 official French runs. The Farhan team used GPT-4 and GPT-4o. The separate Spanish comparison used 2023 training references.

Released
November 2023
Kind
Translation
Level
University
Scored
Checked automatically
Answers
Open-ended
Measures
Capability
Language
English, French
In and out
Text → Text
Interaction
Single turn
Items from
Translated
Run by
A host
17
Entries in its paper, 0 models we track

Arampatzis_GoogleTranslate tops the leaderboard with 91.8

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

  1. 1 Arampatzis_GoogleTranslate 376 translated texts 91.8
  2. 2 Frane_TranslationModel 92 translated texts 91.8
  3. 3 Dajana&Kathy_TranslationModel 376 translated texts 91.2
  4. 4 UBO_SDL 312 translated texts 90.2
  5. 5 Arampatzis_MarianMT 376 translated texts 90.0
  6. 6 Tomislav&Rowan_MarianMTModel 376 translated texts 90.0
  7. 7 UBO_ChatGPT 312 translated texts 89.2
  8. 8 UBO_DeepL 312 translated texts 89.2
  9. 9 UAms_T5-base_ft 376 translated texts 89.0
  10. 10 Arampatzis_mBART 376 translated texts 88.1
  11. 11 Arampatzis_M2M100 376 translated texts 87.7
  12. 12 UAms_Marian_ft 376 translated texts 81.7
  13. 13 Farhan_2 376 translated texts 73.0
  14. 14 Farhan_1 376 translated texts 68.4
  15. 15 jokester_MarianMTModel 49 translated texts 66.8
  16. 16 Arampatzis_opus_mt 63 translated texts 66.5
  17. 17 Arampatzis_T5 63 translated texts 65.3

Source: The CLEF JOKER 2024 Task 3: Pun Translation paper

#EntryScoreSettingDateSource
1 Arampatzis_GoogleTranslate 91.8 376 translated texts Official leaderboard
2 Frane_TranslationModel 91.8 92 translated texts Official leaderboard
3 Dajana&Kathy_TranslationModel 91.2 376 translated texts Official leaderboard
4 UBO_SDL 90.2 312 translated texts Official leaderboard
5 Arampatzis_MarianMT 90.0 376 translated texts Official leaderboard
6 Tomislav&Rowan_MarianMTModel 90.0 376 translated texts Official leaderboard
7 UBO_ChatGPT 89.2 312 translated texts Official leaderboard
8 UBO_DeepL 89.2 312 translated texts Official leaderboard
9 UAms_T5-base_ft 89.0 376 translated texts Official leaderboard
10 Arampatzis_mBART 88.1 376 translated texts Official leaderboard
11 Arampatzis_M2M100 87.7 376 translated texts Official leaderboard
12 UAms_Marian_ft 81.7 376 translated texts Official leaderboard
13 Farhan_2 73.0 376 translated texts Official leaderboard
14 Farhan_1 68.4 376 translated texts Official leaderboard
15 jokester_MarianMTModel 66.8 49 translated texts Official leaderboard
16 Arampatzis_opus_mt 66.5 63 translated texts Official leaderboard
17 Arampatzis_T5 65.3 63 translated texts Official leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

Versions

More in Multilingual

All 545 in Multilingual
  • MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
  • WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET. K-EXAONE 90.5%
  • MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language. Qwen3.7-Max 87%
  • CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories. Solar Open 2 90.7%
  • Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy. Claude Opus 5 95.7%
  • MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%

Data audit

7 of 8 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the benchmark’s paper (17 entries).
  • Pass Models scored 17 entries on its leaderboard, none a model we track.
  • Note Newest score No scores yet.
  • Note Frontier models No models scored yet.
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on ceur-ws.org: 17 entries at the last read, none a model we track.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check