LLM如何看待创造力：具有可解释推理的视觉创造力零样本评分
How LLMs See Creativity: Zero-Shot Scoring of Visual Creativity with Interpretable Reasoning
AI总结：
研究多模态大模型能否零样本评估视觉创造力，并分析其推理过程的可解释性。实验表明模型评分与人类高度一致，但推理并未提升评分准确性。
摘要：
评估视觉图像的原创新颖性对创造力评估提出了持久挑战。使用AI模型进行自动评分在语言领域已被证明有效，但关于评估视觉创造力以及理解模型如何得出评分的关键问题仍然存在。本研究探讨多模态大语言模型（LLMs）是否能在零样本（无需任何微调或人类评分示例）下充当视觉创造力的评判者，以及它们的“推理”输出是否为其评估过程提供了可解释的窗口。我们测试了六个多模态LLM（Gemini 3 Flash、Gemma 4 31B IT、GPT-5.4 Mini、GLM-5v Turbo、Kimi K2.5和Qwen 3.6 Plus）在992张AI生成图像（基于人类编写的提示）和1,500张手绘草图上的表现，这些图像和草图已由人类评分者进行创造力评分。在研究1中，所有模型在两个数据集上都与人类创造力评分表现出显著一致性（AI生成图像上r = .57-.68；草图上r = .29-.68）。在研究2中，我们分析了三个LLM评估相同图像和绘图的逐步推理过程。尽管推理使模型评估变得可解释——展示了它们关注什么、如何平衡原创性与质量、以及如何证明其评分的合理性——但推理并未提高与人类评分的一致性。总之，我们的发现表明，多模态LLM无需额外训练即可匹配人类对视觉创造力的判断，并且它们的推理揭示了AI模型如何评估创造力。实现此流程的开放评分应用程序可在此URL获取。
英文摘要：
Evaluating the originality of visual images poses enduring challenges for creativity assessment. Automated scoring using AI models has proven effective in the verbal domain, yet key questions remain about evaluating visual creativity and understanding how models arrive at their ratings. The present research asks whether multimodal large language models (LLMs) can serve as judges of visual creativity zero-shot (without any fine-tuning or examples of human ratings) and whether their "reasoning" output offers an interpretable window into their evaluation process. We tested six multimodal LLMs (Gemini 3 Flash, Gemma 4 31B IT, GPT-5.4 Mini, GLM-5v Turbo, Kimi K2.5, and Qwen 3.6 Plus) on 992 AI-generated images (based on human-written prompts) and 1,500 hand-drawn sketches scored for creativity by human raters. In Study 1, all models showed substantial alignment with human creativity ratings on both datasets (r = .57-.68 on AI-generated images; r = .29-68 on sketches). In Study 2, we analyzed the step-by-step reasoning processes of three LLMs evaluating the same images and drawings. Although reasoning made model evaluations interpretable -- showing what they attend to, how they balance originality vs. quality, and how they justify their ratings -- reasoning did not improve alignment with human ratings. In sum, our findings indicate that multimodal LLMs can match human judgments of visual creativity without any additional training, and that their reasoning reveals how AI models evaluate creativity. An open scoring app implementing this pipeline is available at https://review-visual-eval-scoring.hf.space.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
