Cards Against LLMs: Benchmarking Humor Alignment in Large Language Models
Humor is one of the most culturally embedded and socially significant dimensions of human communication, yet it remains largely unexplored as a dimension of Large Language Model (LLM) alignment. In this study, five frontier language models play the same Cards Against Humanity games (CAH) as human players. The models select the funniest response from a slate of ten candidate cards across 9,894 rounds. While all models exceed the random baseline, alignment with human preference remains modest. More striking is that models agree with each other substantially more often than they agree with humans. We show that this preference is partly explained by systematic position biases and content preferences, raising the question whether LLM humor judgment reflects genuine preference or structural artifacts of inference and alignment.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Cards Against AI: Predicting Humor in a Fill-in-the-blank Party Game
Humor is an inherently social phenomenon, with humorous utterances shaped by what is socially and culturally accepted. Understanding humor is an important NLP challenge, with many applications to human-computer interacti…
Feature ImportanceHumorReject: Decoupling LLM Safety from Refusal Prefix via A Little Humor
Large Language Models (LLMs) commonly rely on explicit refusal prefixes for safety, making them vulnerable to prefix injection attacks. We introduce HumorReject, a novel data-driven approach that reimagines LLM safety by…
Chumor 2.0: Towards Benchmarking Chinese Humor Understanding
Existing humor datasets and evaluations predominantly focus on English, leaving limited resources for culturally nuanced humor in non-English languages like Chinese. To address this gap, we construct Chumor, the first Ch…
BenchmarkingChumor 1.0: A Truly Funny and Challenging Chinese Humor Understanding Dataset from Ruo Zhi Ba
Existing humor datasets and evaluations predominantly focus on English, lacking resources for culturally nuanced humor in non-English languages like Chinese. To address this gap, we construct Chumor, a dataset sourced fr…
Computational Humor with Multimodal LLMs: Methods, Datasets, Evaluation, and Challenges
Multimodal humor in memes, cartoons, and comics remains difficult for AI systems because intended meaning depends on non-literal mechanisms, shared cultural knowledge, and communicative intent rather than literal scene d…