CLEF JOKER 2026 Task 2: Pun Translation
Models translate English puns into French. The automatic score checks whether a translation contains an annotated French word or phrase that carries the pun.
The test contains 2,656 English puns. Translators separately judged whether selected outputs preserved both the meaning and the wordplay.
- Made by
- CLEF JOKER organizers
- Released
- October 2025
- Kind
- Translation
- Level
- University
- Scored
- Checked automatically
- Answers
- Open-ended
- Tasks
- Public, plus a private test set
- Measures
- Capability
- Subject
- Languages and literature
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- A host
ensemble_0_0_0_100_gpt_pooled_transl tops the leaderboard with 32.6%
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.
- 32.6 ensemble_0_0_0_100_gpt_pooled_transl
- 32.2 gpt_5.5_single
- 31.1 ensemble_0_0_0_100_all_pooled_transl
- 29.7 ensemble_0_0_0_100_claude_pooled_transl
- 28.9 ensemble_0_0_0_100_gemini_pooled_transl
- 28.7 Satir_CVO_Direct_mistral-large:123b
- 27.9 QANCHOR_TOP25
- 27.9 GROQ_FULL_PARTIAL_5
- 27.9 NOTORCH_SELECTOR_MICRO5
- 27.9 NOTORCH_SELECTOR_STRICT
- 27.9 QANCHOR_REFONLY
- 25.8 Satir_Baseline
- 25.2 UAms_marian-ft_content0.0
- 25.2 UAms_marian-ft_content0.25
- 25.1 Satir_BaselineCVO
- 24.8 Satir_Classifier
- 24.6 UAms_MarianMT_XLMRoBERTa
- 24.5 UAms_opus-mt-en-fr-finetuned
- 23.9 UAms_MarianMT_CamemBERT
- 23.6 UAms_nllb-ft_content0.25
- 22.7 Satir_ClassifierCVO
- 22.3 PixelPair_groq_llama
- 17.3 UAms_mbart-ft_content0.25
- 7.05 identity_run
- 1 ensemble_0_0_0_100_gpt_pooled_transl 32.6%
- 2 gpt_5.5_single 32.2%
- 3 ensemble_0_0_0_100_all_pooled_transl 31.1%
- 4 ensemble_0_0_0_100_claude_pooled_transl 29.7%
- 5 ensemble_0_0_0_100_gemini_pooled_transl 28.9%
- 6 Satir_CVO_Direct_mistral-large:123b 28.7%
- 7 QANCHOR_TOP25 27.9%
- 8 GROQ_FULL_PARTIAL_5 27.9%
- 9 NOTORCH_SELECTOR_MICRO5 27.9%
- 10 NOTORCH_SELECTOR_STRICT 27.9%
- 11 QANCHOR_REFONLY 27.9%
- 12 Satir_Baseline 25.8%
- 13 UAms_marian-ft_content0.0 25.2%
- 14 UAms_marian-ft_content0.25 25.2%
- 15 Satir_BaselineCVO 25.1%
- 16 Satir_Classifier 24.8%
- 17 UAms_MarianMT_XLMRoBERTa 24.6%
- 18 UAms_opus-mt-en-fr-finetuned 24.5%
- 19 UAms_MarianMT_CamemBERT 23.9%
- 20 UAms_nllb-ft_content0.25 23.6%
- 21 Satir_ClassifierCVO 22.7%
- 22 PixelPair_groq_llama 22.3%
- 23 UAms_mbart-ft_content0.25 17.3%
- 24 identity_run 7.05%
Source: The CLEF JOKER 2026 Task 2: Pun Translation paper
| # | Entry | Score | Setting | Submitted by | Date | Source |
|---|---|---|---|---|---|---|
| 1 | ensemble_0_0_0_100_gpt_pooled_transl | 32.6% | dsgt-arc | Official leaderboard | ||
| 2 | gpt_5.5_single | 32.2% | dsgt-arc | Official leaderboard | ||
| 3 | ensemble_0_0_0_100_all_pooled_transl | 31.1% | dsgt-arc | Official leaderboard | ||
| 4 | ensemble_0_0_0_100_claude_pooled_transl | 29.7% | dsgt-arc | Official leaderboard | ||
| 5 | ensemble_0_0_0_100_gemini_pooled_transl | 28.9% | dsgt-arc | Official leaderboard | ||
| 6 | Satir_CVO_Direct_mistral-large:123b | 28.7% | n0shi | Official leaderboard | ||
| 7 | QANCHOR_TOP25 | 27.9% | arampageos | Official leaderboard | ||
| 8 | GROQ_FULL_PARTIAL_5 | 27.9% | arampageos | Official leaderboard | ||
| 9 | NOTORCH_SELECTOR_MICRO5 | 27.9% | arampageos | Official leaderboard | ||
| 10 | NOTORCH_SELECTOR_STRICT | 27.9% | arampageos | Official leaderboard | ||
| 11 | QANCHOR_REFONLY | 27.9% | arampageos | Official leaderboard | ||
| 12 | Satir_Baseline | 25.8% | n0shi | Official leaderboard | ||
| 13 | UAms_marian-ft_content0.0 | 25.2% | ezrasmink | Official leaderboard | ||
| 14 | UAms_marian-ft_content0.25 | 25.2% | ezrasmink | Official leaderboard | ||
| 15 | Satir_BaselineCVO | 25.1% | n0shi | Official leaderboard | ||
| 16 | Satir_Classifier | 24.8% | n0shi | Official leaderboard | ||
| 17 | UAms_MarianMT_XLMRoBERTa | 24.6% | robinflier | Official leaderboard | ||
| 18 | UAms_opus-mt-en-fr-finetuned | 24.5% | ezrasmink | Official leaderboard | ||
| 19 | UAms_MarianMT_CamemBERT | 23.9% | robinflier | Official leaderboard | ||
| 20 | UAms_nllb-ft_content0.25 | 23.6% | ezrasmink | Official leaderboard | ||
| 21 | Satir_ClassifierCVO | 22.7% | n0shi | Official leaderboard | ||
| 22 | PixelPair_groq_llama | 22.3% | madhangi | Official leaderboard | ||
| 23 | UAms_mbart-ft_content0.25 | 17.3% | ezrasmink | Official leaderboard | ||
| 24 | identity_run | 7.05% | Baseline | 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
- 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
All 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access Public tasks, plus a private test set.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the benchmark’s paper (24 entries).
- Pass Models scored 24 entries on its leaderboard, none a model we track.
- Note Newest score No scores yet.
- Note Frontier models No models scored yet.
- Note Headroom No scores yet.
Leaderboard import 3 pass
- Pass Importer We read the board on codabench.org: 24 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.