# Bridging the Creativity Understanding Gap: Small-Scale Human Alignment Enables Expert-Level Humor Ranking in LLMs

arXiv: https://arxiv.org/abs/2502.20356 · HTML: https://mlsanity.com/paper/2502.20356/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 20 of its 42 references that are in the map.
- First posted on arXiv 27 Feb 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Human values and AI feedback alignment](https://mlsanity.com/alignment/human-values-and-ai-feedback-alignment/)

## Builds on (most cited first, 20 of 20)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (2022)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigm](https://mlsanity.com/paper/2102.07350/) (2021)
- [From Persona to Personalization: A Survey on Role-Playing Language Agents](https://mlsanity.com/paper/2404.18231/) (2024)
- [Simulating Opinion Dynamics with Networks of LLM-based Agents](https://mlsanity.com/paper/2311.09618/) (2023)
- [Can Machines Learn Morality? The Delphi Experiment](https://mlsanity.com/paper/2110.07574/) (2021)
- [Do Androids Laugh at Electric Sheep? Humor "Understanding" Benchmarks from The New Yorker Caption Contest](https://mlsanity.com/paper/2209.06293/) (2022)
- [Let's Think Outside the Box: Exploring Leap-of-Thought in Large Language Models with Creative Humor Generation](https://mlsanity.com/paper/2312.02439/) (2023)
- [Humor Detection: A Transformer Gets the Last Laugh](https://mlsanity.com/paper/1909.00252/) (2019)
- [The Wisdom of Partisan Crowds: Comparing Collective Intelligence in Humans and LLM-based Agents](https://mlsanity.com/paper/2311.09665/) (2023)
- [Humor in AI: Massive Scale Crowd-Sourced Preferences and Benchmarks for Cartoon Captioning](https://mlsanity.com/paper/2406.10522/) (2024)
- [Humor in Collective Discourse: Unsupervised Funniness Detection in the New Yorker Cartoon Caption Contest](https://mlsanity.com/paper/1506.08126/) (2015)
- [SemEval-2020 Task 7: Assessing Humor in Edited News Headlines](https://mlsanity.com/paper/2008.00304/) (2020)
- [Crowd Score: A Method for the Evaluation of Jokes using Large Language Model AI Voters as Judges](https://mlsanity.com/paper/2212.11214/) (2022)
- [Text Is Not All You Need: Multimodal Prompting Helps LLMs Understand Humor](https://mlsanity.com/paper/2412.05315/) (2024)

## Built on by (most cited first, 12 of 12)

- [Creative Preference Optimization](https://mlsanity.com/paper/2505.14442/) (2025)
- [Which LLMs Get the Joke? Probing Non-STEM Reasoning Abilities with HumorBench](https://mlsanity.com/paper/2507.21476/) (2025)
- [ConlangCrafter: Constructing Languages with a Multi-Hop LLM Pipeline](https://mlsanity.com/paper/2508.06094/) (2025)
- [Improving Task Diversity in Label Efficient Supervised Finetuning of LLMs](https://mlsanity.com/paper/2507.21482/) (2025)
- [Uncovering the Computational Ingredients of Human-Like Representations in LLMs](https://mlsanity.com/paper/2510.01030/) (2025)
- [Engagement Undermines Safety: How Stereotypes and Toxicity Shape Humor in Language Models](https://mlsanity.com/paper/2510.18454/) (2025)
- [Cards Against LLMs: Benchmarking Humor Alignment in Large Language Models](https://mlsanity.com/paper/2604.08757/) (2026)
- [Learning to Think Like a Cartoon Captionist: Incongruity-Resolution Supervision for Multimodal Humor Understanding](https://mlsanity.com/paper/2604.15210/) (2026)
- [Culture-Aware Humorous Captioning: Multimodal Humor Generation across Cultural Contexts](https://mlsanity.com/paper/2604.18091/) (2026)
- [HumorRank: A Tournament-Based Leaderboard for Evaluating Humor Generation in Large Language Models](https://mlsanity.com/paper/2604.19786/) (2026)
- [AGC-Bench: Measuring Artificial General Creativity](https://mlsanity.com/paper/2607.01152/) (2026)
- [Towards High-Level Semantic Intelligence](https://mlsanity.com/paper/2607.24082/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have shown significant limitations in understanding creative content, as demonstrated by Hessel et al. (2023)'s influential work on the New Yorker Cartoon Caption Contest (NYCCC). Their study exposed a substantial gap between LLMs and humans in humor comprehension, establishing that understanding and evaluating creative content is key challenge in AI development. We revisit this challenge by decomposing humor understanding into three components and systematically improve each: enhancing visual understanding through improved annotation, utilizing LLM-generated humor reasoning and explanations, and implementing targeted alignment with human preference data. Our refined approach achieves 82.4% accuracy in caption ranking, singificantly improving upon the previous 67% benchmark and matching the performance of world-renowned human experts in this domain. Notably, while attempts to mimic subgroup preferences through various persona prompts showed minimal impact, model finetuning with crowd preferences proved remarkably effective. These findings reveal that LLM limitations in creative judgment can be effectively addressed through focused alignment to specific subgroups and individuals. Lastly, we propose the position that achieving artificial general intelligence necessitates systematic collection of human preference data across creative domains. We advocate that just as human creativity is deeply influenced by individual and cultural preferences, training LLMs with diverse human preference data may be essential for developing true creative understanding.