EmoGenie: Benchmarking and Evaluating Expressed and Evoked Emotion in T2I Generation with MLLMs

📄 arXiv: 2610.18455v1 📥 PDF

作者: Zixuan Xing, Lancheng Gao, Shengyan Li, Huiyu Duan, Zhaolin Cai, Guangtao Zhai, Xiongkuo Min

分类: cs.CV

发布日期: 2026-10-13

备注: Comments: Corresponding author: Xiongkuo Min. Equal contribution: Zixuan Xing and Lancheng Gao

💡 一句话要点

提出EmoGenie基准以评估T2I生成图像中的情感表达与唤起

🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)

关键词: 情感计算 文本到图像生成 多模态大语言模型 AI生成内容 情感评估 基准测试 人机交互

📋 核心要点

  1. 现有的情感基准主要基于自然图像，缺乏针对生成图像的统一基准，导致情感表达与唤起的研究不足。
  2. 本文提出EmoGenie基准，包含6400幅生成图像及其情感标注，旨在评估T2I生成和MLLM理解的能力。
  3. 实验表明，EmoProbe在情感问答任务中表现优越，其情感和美学预测与人类评分的相关性显著提高。

📝 摘要（中文）

随着人工智能生成内容（AIGC）的快速发展，文本到图像（T2I）模型在生成具有特定情感意图的图像方面得到了广泛应用。然而，生成图像所传达的情感效果常常偏离目标预期，且现有的多模态大语言模型（MLLMs）在解读AI生成视觉内容的情感方面的可靠性尚不明确。为此，本文提出了EmoGenie，一个用于评估AI生成图像情感的基准，涵盖了6400幅由八个T2I模型生成的图像，配以800个情感目标提示，并通过问题-答案对（EmoGenie-QA）和连续人类评分（EmoGenie-Assess）进行注释。实验结果表明，EmoProbe在情感相关的问答任务中表现优越，其情感和美学预测与人类评分的相关性更高，展示了其在AI生成图像的统一粗到细情感理解中的先进能力。

🔬 方法详解

问题定义：本文旨在解决现有情感基准缺乏针对AI生成图像的统一评估工具的问题，现有方法在情感表达与唤起的研究中存在不足。

核心思路：提出EmoGenie基准，通过生成6400幅图像并进行情感标注，结合问答对和人类评分，来全面评估生成图像的情感表现。

技术框架：整体架构包括图像生成模块、情感标注模块（EmoGenie-QA和EmoGenie-Assess）以及EmoProbe模型，用于情感问答和评分预测。

关键创新：EmoGenie基准的建立和EmoProbe模型的开发是本研究的主要创新，前者提供了系统的评估工具，后者在情感理解上实现了从粗到细的统一处理。

关键设计：EmoGenie-QA包含90,000个封闭式问题和1,057个开放式问题，EmoGenie-Assess通过13,175个平均意见分数（MOS）进行情感和美学预测，EmoProbe结合了响应对齐和维度特定的回归头。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果显示，EmoProbe在情感问答任务中取得了显著的性能提升，其情感和美学预测与人类评分的相关性提高，展示了其在情感理解方面的先进能力。具体而言，EmoProbe在多个基准测试中均超越了现有的最先进方法。

🎯 应用场景

该研究的潜在应用领域包括情感计算、AI艺术创作和人机交互等。通过提供一个统一的情感评估基准，EmoGenie可以帮助研究人员和开发者更好地理解和优化AI生成内容的情感表现，推动相关技术的发展与应用。

📄 摘要（原文）

Rapid advances in artificial intelligence-generated content (AIGC) have significantly promoted using text-to-image (T2I) models to generate images with specific affective intentions. However, the emotional effects conveyed by generated images frequently deviate from target expectations, and it remains unclear whether multimodal large language models (MLLMs) can reliably interpret the emotion of AI-generated visual content. Existing affective benchmarks are predominantly built on natural images, while affective studies of generated images have largely focused on emotions expressed in portraits, lacking a unified benchmark for studying the emotions expressed and evoked by generated images. We therefore introduce EmoGenie, a benchmark of emotion for AI-generated images that evaluates capabilities in T2I generation and MLLM understanding across expressed and evoked emotions. It comprises 6,400 images generated by eight T2I models using 800 emotion-targeted prompts, annotated with question-answer (QA) pairs, termed EmoGenie-QA, and continuous human ratings, termed EmoGenie-Assess. EmoGenie-QA contains 90,000 closed-set questions in single-image and image-pair settings, as well as 1,057 open-ended questions for emotion description. EmoGenie-Assess benchmarks continuous affective and aesthetic prediction using 13,175 mean opinion scores (MOSs) aggregated from human ratings in terms of valence, arousal, dominance, and aesthetic quality. We further develop EmoProbe, which combines emotion question answering with response-aligned, dimension-specific regression heads for continuous score prediction. Experimental results show that EmoProbe achieves superior results on emotion-related QA tasks, while its affective and aesthetic predictions correlate better with human ratings, demonstrating its state-of-the-art capabilities of unified coarse-to-fine affective understanding for AI-generated images.