论文研究
arxiv.org ·
HumorGen：通过基于人设的蒸馏实现大语言模型幽默生成的认知协同
该论文提出 Cognitive Synergy Framework，利用 Mixture-of-Thought 部署六种认知人设（如 The Absurdist、The Cynic）为同一提示合成多样化喜剧视角，构建理论驱动的数据集并微调 7B 学生模型。作者比较 DPO 与离线组相对变体 O-GRPO，发现两者均未优于 SFT；但 7B HumorGen 变体显著超过更大的指令微调基线，达到开源权重顶级水平并与前沿闭源系统相当。结论认为，对幽默生成而言，认知驱动的数据策划比对齐算法或模型规模更关键。
AI推荐理由 arXiv 正式论文（v3，2026-09-12 修订），摘要明确给出方法（Cognitive Synergy Framework、Mixture-of-Thought、六种认知人设、7B 学生模型蒸馏）与可验证结论：DPO 与 O-GRPO 均未超过 SFT，而 7B HumorGen 显著优于更大的指令微调基线，并接近前沿闭源系统。新信息在于提出“认知驱动的数据策划比对齐算法或模型规模更关键”这一可检验论断，对 LLM 数据合成与微调路线有参考价值。局限：仅摘要可见，未提供具体评测基准、数据集规模与人工评估细节，幽默生成属较窄任务，结论外推需谨慎。
72
正文
Computer Science > Computation and Language
arXiv:2604.09629 (cs)
[Submitted on 19 Mar 2026 (v1), last revised 12 Sep 2026 (this version, v3)]
Title:HumorGen: Cognitive Synergy for Humor Generation in Large Language Models via Persona-Based Distillation
Authors:Edward Ajayi, Prasenjit Mitra
View a PDF of the paper titled HumorGen: Cognitive Synergy for Humor Generation in Large Language Models via Persona-Based Distillation, by Edward Ajayi and Prasenjit Mitra
View PDF
HTML (experimental)
Abstract:Humor generation poses a significant challenge for Large Language Models (LLMs), because their standard training objective (next-token prediction) inherently conflicts with the surprise and incongruity required for comedy. To bridge this gap, we introduce the Cognitive Synergy Framework, a methodology for generating highquality humor data inspired by psychological theories of humor. Utilizing a Mixtureof-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 produces a theory-grounded dataset, which we use to fine-tune a 7B-parameter student model. We further evaluate two alignment strategies, Direct Preference Optimization (DPO) and an offline group-relative variant O-GRPO, finding that neither improves over SFT. However, our 7B HumorGen model variants significantly outperform larger instruction-tuned baselines and achieve top-tier open-weight performance while remaining competitive with frontier proprietary systems. These results suggest that cognitively driven data curation is more critical than alignment algorithms or model scale for humor generation.
Comments:
Subjects:
Computation and Language (cs.CL)
Cite as:
arXiv:2604.09629 [cs.CL]
(or
arXiv:2604.09629v3 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2604.09629
Focus to learn more
arXiv-issued DOI via DataCite
Submission history
From: Edward Ajayi [view email]
[v1]
Thu, 19 Mar 2026 13:12:53 UTC (1,727 KB)
[v2]
Wed, 27 May 2026 20:51:39 UTC (3,868 KB)
[v3]
Sat, 12 Sep 2026 17:24:27 UTC (3,993 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled HumorGen: Cognitive Synergy for Humor Generation in Large Language Models via Persona-Based Distillation, by Edward Ajayi and Prasenjit Mitra
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CL
new
|
recent
| 2026-04
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
export BibTeX citation
Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? |
Disable MathJax (What is MathJax?)