HumorGen: Cognitive Synergy for Humor Generation in Large Language Models via Persona-Based Distillation

• URL: http://arxiv.org/abs/2604.09629v1
• Date: Thu, 19 Mar 2026 13:12:53 GMT
• Title: HumorGen: Cognitive Synergy for Humor Generation in Large Language Models via Persona-Based Distillation
• Abstract summary: We introduce the Cognitive Synergy Framework, a methodology for generating high-quality humor data.<n>We deploy six cognitive personas to synthesize diverse comedic perspectives for a given prompt.<n>We find that cognitive-driven data is far more critical than alignment algorithms or model scale for humor generation.
• Score: 13.570863502498336
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Humor generation poses a significant challenge for Large Language Models (LLMs), because their standard training objective - predicting the most likely next word - inherently conflicts with the surprise and incongruity needed for comedy. To bridge this gap, we introduce the Cognitive Synergy Framework, a theoretically grounded methodology for generating high-quality humor data inspired by psychological theories of humor. Utilizing a Mixture-of-Thought (MoT) approach, we deploy six cognitive personas (e.g., The Absurdist, The Cynic) to synthesize diverse comedic perspectives for a given prompt. This framework creates a theoretically grounded dataset, which we use to fine-tune a 7B-parameter student model. We compare Direct Preference Optimization (DPO) and a novel Offline Group Relative Policy Optimization (O-GRPO); our 7B model significantly outperforms larger instruction-tuned baselines and achieves performance competitive with state-of-the-art proprietary models. We find that cognitive-driven data curation is far more critical than alignment algorithms or model scale for humor generation. Code and data will be available upon publication.

Related papers

• HumorRank: A Tournament-Based Leaderboard for Evaluating Humor Generation in Large Language Models [13.570863502498336]
  We introduce HumorRank, a tournament-based evaluation framework and leaderboard for textual humor generation.<n>Using SemEval-2026 MWAHAHA test dataset, we conduct an automated pairwise evaluation across nine models spanning proprietary, open-weight, and specialized systems.<n>Our results demonstrate that HumorRank yields statistically grounded model stratifications, showing that humor quality is driven by mastery of comedic mechanisms rather than model scale alone.
  arXiv  Detail & Related papers  (2026-03-31T18:54:15Z)
• From Perception to Punchline: Empowering VLM with the Art of In-the-wild Meme [5.462301274468853]
  We propose HUMOR, a novel framework that guides memes through hierarchical reasoning and aligns them with group-wise human preferences.<n>To capture subjective humor, we train a pairwise reward model that operates within groups of memes sharing the same template.<n>Our work presents a general training paradigm for open-ended, human-aligned multimodal generation.
  arXiv  Detail & Related papers  (2025-12-31T01:35:49Z)
• PRISM: A Personality-Driven Multi-Agent Framework for Social Media Simulation [24.36529339525981]
  We introduce the Personality-Refracted Intelligent Simulation Model (PRISM) for continuous emotional evolution with a personalityconditional partially observable Markov decision process.<n>PRISM achieves superior consistency with human ground truth, outperforming standard homogeneous and Big Five benchmarks.<n>This framework effectively replicates emergent phenomena such as rational suppression and affective resonance, offering a robust tool for analyzing complex social media ecosystems.
  arXiv  Detail & Related papers  (2025-12-22T23:31:49Z)
• HUMORCHAIN: Theory-Guided Multi-Stage Reasoning for Interpretable Multimodal Humor Generation [13.49193658655368]
  Humor, as both a creative human activity and a social binding mechanism, has long posed a major challenge for AI generation.<n>We propose HUMORCHAIN, a theory-guided multi-stage reasoning framework.<n>It integrates visual semantic parsing, humor- and psychology-based reasoning, and a fine-tuned discriminator for humor evaluation.
  arXiv  Detail & Related papers  (2025-11-21T09:52:46Z)
• A Survey on Generative Recommendation: Data, Model, and Tasks [55.36322811257545]
  generative recommendation reconceptualizes recommendation as a generation task rather than discriminative scoring.<n>This survey provides a comprehensive examination through a unified tripartite framework spanning data, model, and task dimensions.<n>We identify five key advantages: world knowledge integration, natural language understanding, reasoning capabilities, scaling laws, and creative generation.
  arXiv  Detail & Related papers  (2025-10-31T04:02:58Z)
• Prismatic Synthesis: Gradient-based Data Diversification Boosts Generalization in LLM Reasoning [77.120955854093]
  We show that data diversity can be a strong predictor of generalization in language models.<n>We introduce G-Vendi, a metric that quantifies diversity via the entropy of model-induced gradients.<n>We present Prismatic Synthesis, a framework for generating diverse synthetic data.
  arXiv  Detail & Related papers  (2025-05-26T16:05:10Z)
• MindGYM: What Matters in Question Synthesis for Thinking-Centric Fine-Tuning? [51.85759493254735]
  MindGYM is a structured and scalable framework for question synthesis.<n>It infuses high-level reasoning objectives to shape the model's synthesis behavior.<n>It composes more complex multi-hop questions based on QA seeds for deeper reasoning.
  arXiv  Detail & Related papers  (2025-03-12T16:03:03Z)
• Explore Theory of Mind: Program-guided adversarial data generation for theory of mind reasoning [88.68573198200698]
  We introduce ExploreToM, the first framework to allow large-scale generation of diverse and challenging theory of mind data.<n>Our approach leverages an A* search over a custom domain-specific language to produce complex story structures and novel, diverse, yet plausible scenarios.<n>Our evaluation reveals that state-of-the-art LLMs, such as Llama-3.1-70B and GPT-4o, show accuracies as low as 0% and 9% on ExploreToM-generated data.
  arXiv  Detail & Related papers  (2024-12-12T21:29:00Z)
• Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines [74.42485647685272]
  We focus on Generative Masked Language Models (GMLMs) We train a model to fit conditional probabilities of the data distribution via masking, which are subsequently used as inputs to a Markov Chain to draw samples from the model. We adapt the T5 model for iteratively-refined parallel decoding, achieving 2-3x speedup in machine translation with minimal sacrifice in quality.
  arXiv  Detail & Related papers  (2024-07-22T18:00:00Z)
• Towards Multimodal Prediction of Spontaneous Humour: A Novel Dataset and First Results [84.37263300062597]
  Humor is a substantial element of human social behavior, affect, and cognition. Current methods of humor detection have been exclusively based on staged data, making them inadequate for "real-world" applications. We contribute to addressing this deficiency by introducing the novel Passau-Spontaneous Football Coach Humor dataset, comprising about 11 hours of recordings.
  arXiv  Detail & Related papers  (2022-09-28T17:36:47Z)

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.