New Yorker Caption Contest Humor Understanding
Models match a cartoon to its caption, choose the funnier of two captions, and explain a joke. They receive the cartoon or its description.
The authors report matching and ranking accuracy across five splits. Their Hugging Face dataset has 78 likes.
- Made by
- Allen Institute for AI and collaborators
- Released
- September 2022
- Kind
- General visual questions
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- Public, plus a private test set
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Real work and use, Written by crowdworkers
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Last model added
- October 2019
- Scores refreshed
- November 20th, 2026
GPT-4 (snapshot unspecified) tops the leaderboard with 84.5% T5-11B leads with 70.8%
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The paper reports no overall score, so this headline is the matching accuracy column, which is the first task. 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. The best model we track is T5-11B, at 70.8%. This view shows only models we track, each at its best.
- 1 GPT-4 (snapshot unspecified) 5-shot 84.5%
- 2 GPT3-175B finetuned 75.1%
- 3
T5-11B from pixels 70.8%
- 4 GPT 3.5 (snapshot unspecified) 5-shot 63.8%
- 5 CLIP ViT-L/14@336px finetuned 62.3%
- 6 T5-Large from pixels 59.6%
- 7 Caption Only (T5-11B) 19.4%
Source: The New Yorker Caption Contest Humor Understanding paper
- 1
T5-11B from pixels 70.8%
- 2 T5-Large from pixels 59.6%
Source: The New Yorker Caption Contest Humor Understanding paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4 (snapshot unspecified) | 84.5% | 5-shot | Official leaderboard | |
| 2 | GPT3-175B | 75.1% | finetuned | Official leaderboard | |
| 3 | | 70.8% | from pixels | Oct 2019 | Official leaderboard |
| 4 | GPT 3.5 (snapshot unspecified) | 63.8% | 5-shot | Official leaderboard | |
| 5 | CLIP ViT-L/14@336px | 62.3% | finetuned | Official leaderboard | |
| 6 | T5-Large | 59.6% | from pixels | Official leaderboard | |
| 7 | Caption Only (T5-11B) | 19.4% | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official New Yorker Caption Contest Humor Understanding 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.
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy.
Qwen 3.6 Plus 86%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy.
Seed2.1 Pro 90.7%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy.
Qwen 3.8 Max 97.7%
- MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy.
Qwen 3.8 Max 85.9%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 years ago, in October 2019, for T5-11B. Every benchmark
Catalog record 2 pass
- Pass Task access Public tasks, plus a private test set.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (7 entries).
- Pass Models scored 2 models we track, plus 5 other entries on its leaderboard.
- Warning Newest score The newest came 7 years ago, in October 2019, for T5-11B.
- Note Frontier models None of its models are on the capabilities index.
- Pass Headroom The best is 70.8%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on aclanthology.org: 2 rows for models we track and 5 other entries at the last read.
- 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.