📄 中文摘要
表情包作为现代网络交流的重要元素，不仅是静态的艺术品，还可以作为对话中的互动回复。尽管计算研究已关注表情包的内在特性，但表情包在动态和上下文中创造幽默的使用仍然是网络科学中一个未被充分研究的领域。为了解决这一空白，提出了表情包回复选择任务，并呈现了MaMe-Re（漫画表情包回复基准），这是一个包含100,000对人类标注的开放许可日本漫画面板和社交媒体帖子的数据集。分析结果揭示了三个关键见解：大型语言模型（LLMs）在捕捉复杂社会语境方面显示出初步证据。
作为回复的表情包：模型能否选择幽默的漫画面板回应？
出处: Memes-as-Replies: Can Models Select Humorous Manga Panel Responses?
发布: 2026年2月19日
📄 English Summary
Memes-as-Replies: Can Models Select Humorous Manga Panel Responses?
Memes are a significant element of modern web communication, serving not only as static artifacts but also as interactive replies in conversations. While computational research has focused on the intrinsic properties of memes, the dynamic and contextual use of memes to create humor remains an understudied area in web science. To address this gap, the Meme Reply Selection task is introduced, along with the MaMe-Re (Manga Meme Reply Benchmark), a dataset comprising 100,000 human-annotated pairs of openly licensed Japanese manga panels and social media posts. The analysis reveals three key insights: large language models (LLMs) show preliminary evidence of capturing complex sociocultural contexts.