Computational Humor with Multimodal LLMs: Methods, Datasets, Evaluation, and Challenges

  • 类型:arxiv
  • 标识:2607.19011
  • 链接:https://arxiv.org/abs/2607.19011
  • 主分类:multimodal
  • 形态:survey
  • 被引:0
  • 被引来源:Semantic Scholar + OpenAlex
  • S2被引:0
  • OpenAlex被引:0
  • 影响力被引:0
  • TLDR:This survey focuses on visual humor understanding in single-image and multi-panel artifacts, while treating humor generation as an emerging downstream frontier, and synthesizes benchmark design, evaluation protocols, and modeling paradigms based on multimodal alignment, evidence-grounded reasoning, and controlled generation.
  • OpenAlex ID:W7170090039
  • OpenAlex DOI:10.48550/arxiv.2607.19011
  • DOI:10.48550/arxiv.2607.19011
  • DOI来源:OpenAlex
  • 开放获取:green
  • 开放获取链接:https://doi.org/10.48550/arxiv.2607.19011
  • OpenAlex更新:2026-08-24
  • 副分类:evaluation
  • 待LLM分类:否
  • 标题中文:多模态 LLM 的计算幽默:方法、数据集、评估与挑战
  • TLDR中文:本综述聚焦于单图与多格视觉作品中的幽默理解,同时将幽默生成视为新兴的下游前沿方向,并围绕多模态对齐、证据 grounded 推理与可控生成,对基准设计、评估协议与建模范式进行系统综述。
  • 来源文件
  • /inbox/tom/_candidates/2026-07-23-agent-rag-longcontext-candidates.json
  • [S2 enrich]
  • [OpenAlex backfill]