# lmfaoooo at SemEval-2026 Task 1: Humor Is an Audience. Preference Modeling for Constrained Humor Generation

arXiv: https://arxiv.org/abs/2606.00022 · HTML: https://mlsanity.com/paper/2606.00022/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 36 references that are in the map.
- First posted on arXiv 14 Apr 2026.

## Builds on (most cited first, 11 of 11)

- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversity](https://mlsanity.com/paper/2510.01171/) (2025)
- [RankME: Reliable Human Ratings for Natural Language Generation](https://mlsanity.com/paper/1803.05928/) (2018)
- [Humor Detection: A Transformer Gets the Last Laugh](https://mlsanity.com/paper/1909.00252/) (2019)
- [ExPUNations: Augmenting Puns with Keywords and Explanations](https://mlsanity.com/paper/2210.13513/) (2022)
- [Innovative Thinking, Infinite Humor: Humor Research of Large Language Models through Structured Thought Leaps](https://mlsanity.com/paper/2410.10370/) (2024)
- [Paranoid Transformer: Reading Narrative of Madness as Computational Approach to Creativity](https://mlsanity.com/paper/2007.06290/) (2020)
- [Humor Mechanics: Advancing Humor Generation with Multistep Reasoning](https://mlsanity.com/paper/2405.07280/) (2024)
- [Post Turing: Mapping the landscape of LLM Evaluation](https://mlsanity.com/paper/2311.02049/) (2023)
- [Reliable, Reproducible, and Really Fast Leaderboards with Evalica](https://mlsanity.com/paper/2412.11314/) (2024)
- [Who Laughs with Whom? Disentangling Influential Factors in Humor Preferences across User Clusters and LLMs](https://mlsanity.com/paper/2601.03103/) (2026)

## Abstract (arXiv)

Humor generation remains difficult not only because producing fluent, novel jokes is hard, but because "funny" is audience-dependent and supervision is noisy -- preferences vary with audience, context, and culture, and annotator agreement is often low. In this paper, we describe our system for the SemEval-2026 Task-1 (MWAHAHA), which focuses on humor generation under explicit constraints. The task evaluates submitted systems via human preference judgments in 1-on-1 arena-style comparisons. We adopt a "generate-many -> select-best" strategy. First, we generate a diverse pool of candidates per instance using multi-step prompting, model ensembling, and diversity-oriented decoding. Second, we select outputs using a preference model that approximates a "reader" by learning from human comparisons rather than absolute funniness scores. To support this approach, we release 2.5K human pairwise judgments collected through the Humor Arena prototype. We further propose an interpretable pipeline that converts labeled comparisons into a preference model. Across three preference datasets, our models consistently outperform baselines and show stronger cross-domain transfer. Finally, we apply the learned preference model to rank candidates for the MWAHAHA setting and release intermediate artifacts (candidate pools and rankings) to facilitate follow-up work. Our system ranked 1st in the English and Chinese subtasks of MWAHAHA and 2nd in the Spanish subtask.