English(EN) Cross-Model Humor Preference Modeling with Cards Against Humanity

研究发现：大型语言模型可以预测其他模型的幽默偏好

作者 PulseAugur 编辑部 · [1 个来源] · 2026-08-11 04:00

一项新的研究论文探讨了是否可以使用类似 Cards Against Humanity 的任务，让一个大型语言模型预测另一个模型的幽默偏好。该研究将 GPT-4o 与 Claude Opus 4.5 进行对比，发现简单的模仿指令只能带来微小的改进，而提供其他模型选择的行为证据，尤其是附带理由，则能显著提高准确性。这表明一种类似心智理论的行为，模型可以根据观察到的偏好调整其响应，而不是内化它们。 AI

影响展示了大型语言模型可以表现出类似心智理论的行为，有望改善代理协作和个性化人工智能交互。

排序理由该集群包含一篇详细介绍大型语言模型能力实验的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍大型语言模型能力实验的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 60 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Victor Winter, Farhan Lakhany · 2026-08-11 04:00

使用 Cards Against Humanity 进行跨模型幽默偏好建模

arXiv:2608.07481v1 Announce Type: cross Abstract: This paper investigates whether one large language model can approximate the humor preferences of another in a controlled Cards Against Humanity-style task. Two models - GPT-4o as Czar and Claude Opus-4.5 as Player - are evaluated…

本文如何被排名

报道来源 [1]

使用 Cards Against Humanity 进行跨模型幽默偏好建模

相关实体

相关话题