# Getting Serious about Humor: Crafting Humor Datasets with Unfunny Large Language Models

arXiv: https://arxiv.org/abs/2403.00794 · HTML: https://mlsanity.com/paper/2403.00794/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 40 references that are in the map.
- First posted on arXiv 23 Feb 2024.

## Builds on (most cited first, 15 of 15)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [NLP Evaluation in trouble: On the Need to Measure LLM Data Contamination for each Benchmark](https://mlsanity.com/paper/2310.18018/) (2023)
- [MEGA: Multilingual Evaluation of Generative AI](https://mlsanity.com/paper/2303.12528/) (2023)
- [Is ChatGPT A Good Translator? Yes With GPT-4 As The Engine](https://mlsanity.com/paper/2301.08745/) (2023)
- [Do Androids Laugh at Electric Sheep? Humor "Understanding" Benchmarks from The New Yorker Caption Contest](https://mlsanity.com/paper/2209.06293/) (2022)
- [Exploiting Asymmetry for Synthetic Training Data Generation: SynthIE and the Case of Information Extraction](https://mlsanity.com/paper/2303.04132/) (2023)
- [L3Cube-HingCorpus and HingBERT: A Code Mixed Hindi-English Dataset and BERT Language Models](https://mlsanity.com/paper/2204.08398/) (2022)
- ["President Vows to Cut <Taxes> Hair": Dataset and Analysis of Creative Text Editing for Humorous Headlines](https://mlsanity.com/paper/1906.00274/) (2019)
- [Reverse-Engineering Satire, or "Paper on Computational Humor Accepted Despite Making Serious Advances"](https://mlsanity.com/paper/1901.03253/) (2019)
- [Humor Detection in English-Hindi Code-Mixed Social Media Content : Corpus and Baseline System](https://mlsanity.com/paper/1806.05513/) (2018)
- [Stimulating Creativity with FunLines: A Case Study of Humor Generation in Headlines](https://mlsanity.com/paper/2002.02031/) (2020)
- [Laughing Heads: Can Transformers Detect What Makes a Sentence Funny?](https://mlsanity.com/paper/2105.09142/) (2021)

## Built on by (most cited first, 9 of 9)

- [Can MLLMs Understand the Deep Implication Behind Chinese Images?](https://mlsanity.com/paper/2410.13854/) (2024)
- [Let Androids Dream of Electric Sheep: A Human-Inspired Image Implication Understanding and Reasoning Framework](https://mlsanity.com/paper/2505.17019/) (2025)
- [On the Wings of Imagination: Conflicting Script-based Multi-role Framework for Humor Caption Generation](https://mlsanity.com/paper/2602.06423/) (2026)
- [Harm or Humor: A Multimodal, Multilingual Benchmark for Overt and Covert Harmful Humor](https://mlsanity.com/paper/2603.17759/) (2026)
- [Learning to Think Like a Cartoon Captionist: Incongruity-Resolution Supervision for Multimodal Humor Understanding](https://mlsanity.com/paper/2604.15210/) (2026)
- [HumorRank: A Tournament-Based Leaderboard for Evaluating Humor Generation in Large Language Models](https://mlsanity.com/paper/2604.19786/) (2026)
- [Grounded Satirical Generation with RAG](https://mlsanity.com/paper/2605.10853/) (2026)
- [RAGthoven at SemEval-2026 Task 1: A Multi-Stage Pipeline Walks Into a Benchmark and Barely Clears the Bar](https://mlsanity.com/paper/2607.13189/) (2026)
- [Towards High-Level Semantic Intelligence](https://mlsanity.com/paper/2607.24082/) (2026)

## Abstract (arXiv)

Humor is a fundamental facet of human cognition and interaction. Yet, despite recent advances in natural language processing, humor detection remains a challenging task that is complicated by the scarcity of datasets that pair humorous texts with similar non-humorous counterparts. In our work, we investigate whether large language models (LLMs), can generate synthetic data for humor detection via editing texts. We benchmark LLMs on an existing human dataset and show that current LLMs display an impressive ability to 'unfun' jokes, as judged by humans and as measured on the downstream task of humor detection. We extend our approach to a code-mixed English-Hindi humor dataset, where we find that GPT-4's synthetic data is highly rated by bilingual annotators and provides challenging adversarial examples for humor classifiers.