futurehuhu/GenAI-Bench
收藏资源简介:
GenAI-Bench是一个基准数据集,旨在通过比较从GenAI-Arena收集的人类偏好,来评估多模态大语言模型(MLLMs)判断AI生成内容质量的能力。换句话说,它评估现有MLLMs作为多模态奖励模型的性能,从这个角度看,GenAI-Bench是一个针对多模态生成模型的奖励基准。数据集经过NSFW过滤器和其他启发式方法过滤现有投票,最终包含1735个图像生成投票、919个图像编辑投票和1069个视频生成投票,用于评估MLLMs在与人类偏好对齐方面的性能。每个任务采用成对比较模板,模型被要求为每对AI生成内容输出4个标签(`A>B`、`B>A`、`A=B=Good`、`A=B=Bad`),然后通过比较模型预测与人类偏好来计算平均准确率。
GenAI-Bench is a benchmark designed to benchmark MLLMs’s ability in judging the quality of AI generative contents by comparing with human preferences collected through our GenAI-Arena. In other words, we are evaluating the capabilities of existing MLLMs as a multimodal reward model, and in this view, GenAI-Bench is a reward-bench for multimodal generative models. We filter existing votes collected via NSFW filter and other heuristics, and then finally resulting in 1735 votes for image generation, 919 votes for image editing, and 1069 votes for video generation, which is used to evaluate the performance of MLLMs on aligning with human preferences. We adopts a pairwise comparison template for each tasks, where the model is asked to output 4 labels for each pair of AI generative contents, which are `A>B`, `B>A`, `A=B=Good`, `A=B=Bad`. We then calculate the average accuracy of the model by comparing the models prediction with the human preference.




