Who Laughs with Whom? Disentangling Influential Factors in Humor Preferences across User Clusters and LLMs

• URL: http://arxiv.org/abs/2601.03103v1
• Date: Tue, 06 Jan 2026 15:33:45 GMT
• Title: Who Laughs with Whom? Disentangling Influential Factors in Humor Preferences across User Clusters and LLMs
• Authors: Soichiro Murakami, Hidetaka Kamigaito, Hiroya Takamura, Manabu Okumura,
• Abstract summary: Humor preferences vary widely across individuals and cultures, complicating the evaluation of humor using large language models (LLMs)<n>In this study, we model heterogeneity in humor preferences in Oogiri, a Japanese creative response game, by clustering users with voting logs and estimating cluster-specific weights over interpretable preference factors using Bradley-Terry-Luce models.
• Score: 53.060893644603844
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Humor preferences vary widely across individuals and cultures, complicating the evaluation of humor using large language models (LLMs). In this study, we model heterogeneity in humor preferences in Oogiri, a Japanese creative response game, by clustering users with voting logs and estimating cluster-specific weights over interpretable preference factors using Bradley-Terry-Luce models. We elicit preference judgments from LLMs by prompting them to select the funnier response and found that user clusters exhibit distinct preference patterns and that the LLM results can resemble those of particular clusters. Finally, we demonstrate that, by persona prompting, LLM preferences can be directed toward a specific cluster. The scripts for data collection and analysis will be released to support reproducibility.

Related papers

• ESMC: MLLM-Based Embedding Selection for Explainable Multiple Clustering [79.69917150582633]
  Multi-modal large language models (MLLMs) can be leveraged to achieve user-driven clustering.<n>Our method first discovers that MLLMs' hidden states of text tokens are strongly related to the corresponding features.<n>We also employ a lightweight clustering head augmented with pseudo-label learning, significantly enhancing clustering accuracy.
  arXiv  Detail & Related papers  (2025-11-30T04:36:51Z)
• Investigating Thematic Patterns and User Preferences in LLM Interactions using BERTopic [4.087884819027264]
  This study applies BERTopic to the lmsys-chat-1m dataset, a multilingual conversational corpus built from head-to-head evaluations of large language models (LLMs)<n>The main objective is uncovering thematic patterns in these conversations and examining their relation to user preferences.<n>We analysed relationships between topics and model preferences to identify trends in model-topic alignment.
  arXiv  Detail & Related papers  (2025-10-08T21:13:44Z)
• CUPID: Evaluating Personalized and Contextualized Alignment of LLMs from Interactions [39.554239954719876]
  CUPID is a benchmark of 756 human-curated interaction session histories.<n>We evaluate 10 open and proprietary Large Language Models (LLMs)<n>Our work highlights the need to advance LLM capabilities for more contextually personalized interactions.
  arXiv  Detail & Related papers  (2025-08-03T09:04:48Z)
• HyPerAlign: Interpretable Personalized LLM Alignment via Hypothesis Generation [24.67727411391369]
  HyPerAlign is an interpretable and sample-efficient hypothesis-driven personalization approach for large language models.<n>We conduct experiments on two different personalization tasks, namely authorship attribution and deliberative alignment.<n>Results demonstrate the superiority of hypothesis-driven personalization compared to preference-based fine-tuning methods.
  arXiv  Detail & Related papers  (2025-04-29T18:01:46Z)
• Agent-Centric Personalized Multiple Clustering with Multi-Modal LLMs [40.38930402847949]
  We propose an agent-centric personalized clustering framework.<n>Agents traverse a relational graph to search for clusters based on user interests.<n>Results show that the proposed method achieves NMI scores of 0.9667 and 0.9481 on the Card Order and Card Suits benchmarks.
  arXiv  Detail & Related papers  (2025-03-28T08:45:15Z)
• ComPO: Community Preferences for Language Model Personalization [122.54846260663922]
  ComPO is a method to personalize preference optimization in language models. We collect and release ComPRed, a question answering dataset with community-level preferences from Reddit.
  arXiv  Detail & Related papers  (2024-10-21T14:02:40Z)
• Diverging Preferences: When do Annotators Disagree and do Models Know? [92.24651142187989]
  We develop a taxonomy of disagreement sources spanning 10 categories across four high-level classes. We find that the majority of disagreements are in opposition with standard reward modeling approaches. We develop methods for identifying diverging preferences to mitigate their influence on evaluation and training.
  arXiv  Detail & Related papers  (2024-10-18T17:32:22Z)
• PersonalLLM: Tailoring LLMs to Individual Preferences [11.717169516971856]
  We present a public benchmark, PersonalLLM, focusing on adapting LLMs to provide maximal benefits for a particular user.<n>We curate open-ended prompts paired with many high-quality answers over which users would be expected to display heterogeneous latent preferences.<n>Our dataset and generated personalities offer an innovative testbed for developing personalization algorithms.
  arXiv  Detail & Related papers  (2024-09-30T13:55:42Z)
• Relative Preference Optimization: Enhancing LLM Alignment through Contrasting Responses across Identical and Diverse Prompts [95.09994361995389]
  Relative Preference Optimization (RPO) is designed to discern between more and less preferred responses derived from both identical and related prompts. RPO has demonstrated a superior ability to align large language models with user preferences and to improve their adaptability during the training process.
  arXiv  Detail & Related papers  (2024-02-12T22:47:57Z)
• Large Language Models Enable Few-Shot Clustering [88.06276828752553]
  We show that large language models can amplify an expert's guidance to enable query-efficient, few-shot semi-supervised text clustering. We find incorporating LLMs in the first two stages can routinely provide significant improvements in cluster quality.
  arXiv  Detail & Related papers  (2023-07-02T09:17:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.