[論文レビュー] Jokeasy: Exploring Human-AI Collaboration in Thematic Joke Generation

Yate Ge, Lin Tian+5 more|arXiv (Cornell University)|Feb 10, 2026

Humor Studies and Applications被引用数 0

ひとこと要約

tldr: Jokeasy presents a search-enabled prototype where a dual-role LLM acts as material scout and prototype writer to support thematic joke creation, evaluated through a qualitative study.

ABSTRACT

Thematic jokes are central to stand-up comedy, sitcoms, and public speaking, where contexts and punchlines rely on fresh material - news, anecdotes, and cultural references that resonate with the audience. Recent advances in Large Language Models (LLMs) have enabled interactive joke generation through conversational interfaces. Although LLMs enable interactive joke generation, ordinary conversational interfaces seldom give creators enough agency, control, or timely access to such source material for constructing context and punchlines. We designed Jokeasy, a search-enabled prototype system that integrates a dual-role LLM agent acting as both a material scout and a prototype writer to support human-AI collaboration in thematic joke writing. Jokeasy provides a visual canvas in which retrieved web content is organized into editable inspiration blocks and developed through a multistage workflow. A qualitative study with 13 hobbyists and 5 expert participants (including professional comedians and HCI/AI specialists) showed that weaving real-time web material into this structured workflow enriches ideation and preserves author agency, while also revealing needs for finer search control, tighter chat-canvas integration, and more flexible visual editing. These insights refine our understanding of AI-assisted humour writing and guide future creative-writing tools.

研究の動機と目的

• Motivate and investigate how AI-assisted tools can support thematic joke generation through real-time web material integration.
• Examine whether a structured, multi-stage workflow with a visual canvas preserves creator agency.
• Evaluate how a dual-role LLM (material scout + prototype writer) affects ideation and collaboration.
• Identify user needs to improve search control, chat-canvas integration, and visual editing in AI-assisted humor tools.

提案手法

• Prototype Jokeasy implements a search-enabled interface where retrieved web content is organized into editable inspiration blocks.
• A dual-role LLM agent serves as both material scout (collects and curates source material) and prototype writer (drafts joke material).
• A visual canvas presents inspiration blocks and supports a multistage joke-writing workflow.
• Participants (13 hobbyists and 5 experts) engage with Jokeasy to ideate and develop thematic jokes.
• A qualitative study analyzes user experiences to extract design implications and collaboration dynamics.

実験結果

リサーチクエスチョン

• RQ1 How does integrating real-time web material within a structured workflow influence ideation in thematic joke writing?
• RQ2 Does the material-scout plus writer AI configuration preserve or enhance author agency during co-creation?
• RQ3 What are the design requirements (search control, chat-canvas integration, visual editing) for effective AI-assisted humor tools?
• RQ4 What are the differences in experiences between hobbyists and professional comedians or HCI/AI specialists when using Jokeasy?

主な発見

• We found that weaving real-time web material into the workflow enriches ideation.
• The approach helps preserve author agency during the creative process.
• Participants highlighted a need for finer search control.
• Users requested tighter integration between chat interface and canvas.
• Users desired more flexible visual editing to support iterative refinement.