text-2-video-human-preferences-kling-v2.1-master
收藏魔搭社区2025-12-05 更新2025-12-06 收录
下载链接：
https://modelscope.cn/datasets/Rapidata/text-2-video-human-preferences-kling-v2.1-master
下载链接
链接失效反馈官方服务：
资源简介：
<style>
.vertical-container {
display: flex;
flex-direction: column;
gap: 60px;
}
.image-container img {
height: 150px; /* Set the desired height */
margin:0;
object-fit: contain; /* Ensures the aspect ratio is maintained */
width: auto; /* Adjust width automatically based on height */
}
.image-container {
display: flex; /* Aligns images side by side */
justify-content: space-around; /* Space them evenly */
align-items: center; /* Align them vertically */
}
.container {
width: 90%;
margin: 0 auto;
}
.text-center {
text-align: center;
}
.score-amount {
margin: 0;
margin-top: 10px;
}
.score-percentage {
font-size: 12px;
font-weight: semi-bold;
}
</style>
# Rapidata Video Generation Kling v2.1 Master Human Preference
<a href="https://www.rapidata.ai">
<img src="https://cdn-uploads.huggingface.co/production/uploads/66f5624c42b853e73e0738eb/jfxR79bOztqaC6_yNNnGU.jpeg" width="300" alt="Dataset visualization">
</a>
<a href="https://huggingface.co/datasets/Rapidata/text-2-image-Rich-Human-Feedback">
</a>
In this dataset, ~60k human responses from ~20k human annotators were collected to evaluate Kling v2.1 Master video generation model on our benchmark. This dataset was collected in roughtly 30 min using the [Rapidata Python API](https://docs.rapidata.ai), accessible to anyone and ideal for large scale data annotation.
Explore our latest model rankings on our [website](https://www.rapidata.ai/benchmark).
If you get value from this dataset and would like to see more in the future, please consider liking it ❤️
# Overview
In this dataset, ~60k human responses from ~20k human annotators were collected to evaluate Kling v2.1 Master video generation model on our benchmark. This dataset was collected in roughtly 30 min using the [Rapidata Python API](https://docs.rapidata.ai), accessible to anyone and ideal for large scale data annotation.
The benchmark data is accessible on [huggingface](https://huggingface.co/datasets/Rapidata/text-2-video-human-preferences) directly.
# Explanation of the colums
The dataset contains paired video comparisons. Each entry includes 'video1' and 'video2' fields, which contain links to downscaled GIFs for easy viewing. The full-resolution videos can be found [here](https://huggingface.co/datasets/Rapidata/text-2-video-human-preferences-kling-v2.1-master/tree/main/Videos)
The weighted_results column contains scores ranging from 0 to 1, representing aggregated user responses. Individual user responses can be found in the detailedResults column.
# Alignment
The alignment score quantifies how well an video matches its prompt. Users were asked: "Which video fits the description better?".
## Examples
<div class="vertical-container">
<div class="container">
<div class="text-center">
<q>Aerial view of synchronized swimmers performing intricate patterns in a crystal-clear lake, their movements fluid and graceful under the soft glow of the morning sun.</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(Score: 86.08%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_91_0.gif" width=500>
</div>
<div>
<h3 class="score-amount">Hunyuan </h3>
<div class="score-percentage">(Score: 13.92%)</div>
<img src="https://assets.rapidata.ai/hunyuan_0091_421.gif" width=500>
</div>
</div>
</div>
<div class="container">
<div class="text-center">
<q>A hyper-realistic view of an astronaut inside a spaceship, gazing out at Earth through a large window. Soft ambient light highlights the control panels, creating a serene yet awe-inspiring atmosphere.</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(Score: 11.26%)</div>
<img src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_59_0.gif" width=500>
</div>
<div>
<h3 class="score-amount">Sora </h3>
<div class="score-percentage">(Score: 88.74%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/sora_0059_0.gif" width=500>
</div>
</div>
</div>
</div>
# Coherence
The coherence score measures whether the generated video is logically consistent and free from artifacts or visual glitches. Without seeing the original prompt, users were asked: "Which video has more glitches and is more likely to be AI generated?"
## Examples
<div class="vertical-container">
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(Glitch Rating: 7.35%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_36_0.gif" width="500" alt="Dataset visualization">
</div>
<div>
<h3 class="score-amount">Sora </h3>
<div class="score-percentage">(Glitch Rating: 92.75%)</div>
<img src="https://assets.rapidata.ai/sora_0036_0.gif" width="500" alt="Dataset visualization">
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(Glitch Rating: 87.97%)</div>
<img src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_37_0.gif" width="500" alt="Dataset visualization">
</div>
<div>
<h3 class="score-amount">Ray 2 </h3>
<div class="score-percentage">(Glitch Rating: 12.03%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/ray2_0037_2.gif" width="500" alt="Dataset visualization">
</div>
</div>
</div>
</div>
# Preference
The preference score reflects how visually appealing participants found each video, independent of the prompt. Users were asked: "Which video do you prefer aesthetically?"
## Examples
<div class="vertical-container">
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(Score: 94.77%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_11_0.gif" width="500" alt="Dataset visualization">
</div>
<div>
<h3 class="score-amount">Pika </h3>
<div class="score-percentage">(Score: 5.23%)</div>
<img src="https://assets.rapidata.ai/pika_0011_2286430682.gif" width="500" alt="Dataset visualization">
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(Score: 14.66%)</div>
<img src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_64_0.gif" width="500" alt="Dataset visualization">
</div>
<div>
<h3 class="score-amount">Seedance 1 Pro </h3>
<div class="score-percentage">(Score: 85.34%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/seedance-1-pro-24-7-25_64_0.gif " width="500" alt="Dataset visualization">
</div>
</div>
</div>
</div>
</br>
# About Rapidata
Rapidata's technology makes collecting human feedback at scale faster and more accessible than ever before. Visit [rapidata.ai](https://www.rapidata.ai/) to learn more about how we're revolutionizing human feedback collection for AI development.
# Other Datasets
We run a benchmark of the major video generation models, the results can be found on our [website](https://www.rapidata.ai/leaderboard/video-models). We rank the models according to their coherence/plausiblity, their aligment with the given prompt and style prefernce. The underlying 2M+ annotations can be found here:
- Link to the [Rich Video Annotation dataset](https://huggingface.co/datasets/Rapidata/text-2-video-Rich-Human-Feedback)
- Link to the [Coherence dataset](https://huggingface.co/datasets/Rapidata/Flux_SD3_MJ_Dalle_Human_Coherence_Dataset)
- Link to the [Text-2-Image Alignment dataset](https://huggingface.co/datasets/Rapidata/Flux_SD3_MJ_Dalle_Human_Alignment_Dataset)
- Link to the [Preference dataset](https://huggingface.co/datasets/Rapidata/700k_Human_Preference_Dataset_FLUX_SD3_MJ_DALLE3)
# Rapidata 视频生成 Kling v2.1 Master 人类偏好数据集
<a href="https://www.rapidata.ai">
<img src="https://cdn-uploads.huggingface.co/production/uploads/66f5624c42b853e73e0738eb/jfxR79bOztqaC6_yNNnGU.jpeg" width="300" alt="Dataset visualization">
</a>
<a href="https://huggingface.co/datasets/Rapidata/text-2-image-Rich-Human-Feedback">
</a>
本数据集收集了约2万名人类标注员的总计约6万条标注反馈，用于在我们的基准测试中评估Kling v2.1 Master视频生成模型。本数据集通过[Rapidata Python API](https://docs.rapidata.ai)耗时约30分钟完成采集，该工具面向所有用户开放，非常适合大规模数据标注任务。
可前往我们的[官方网站](https://www.rapidata.ai/benchmark)查看最新的模型排名。
若您从本数据集获益并希望未来获取更多同类资源，欢迎为数据集点赞❤️
# 概述
本数据集收集了约2万名人类标注员的总计约6万条标注反馈，用于在我们的基准测试中评估Kling v2.1 Master视频生成模型。本数据集通过[Rapidata Python API](https://docs.rapidata.ai)耗时约30分钟完成采集，该工具面向所有用户开放，非常适合大规模数据标注任务。基准测试原始数据可直接在[Hugging Face](https://huggingface.co/datasets/Rapidata/text-2-video-human-preferences)获取。
# 字段说明
本数据集包含成对视频对比样本。每条数据均包含`video1`与`video2`字段，其中存储了用于快速预览的压缩版GIF文件链接。全分辨率视频可通过[此链接](https://huggingface.co/datasets/Rapidata/text-2-video-human-preferences-kling-v2.1-master/tree/main/Videos)获取。
`weighted_results`字段存储取值范围为0至1的分数，代表经过聚合的用户反馈结果。单条用户标注详情可在`detailedResults`字段中查看。
# 对齐性 (Alignment)
对齐性得分用于量化视频与对应提示词的匹配程度。标注员被要求回答："哪一段视频更贴合给定的文本描述？"
## 示例
<div class="vertical-container">
<div class="container">
<div class="text-center">
<q>澄澈湖面上，花样游泳运动员同步完成复杂队形，清晨柔光下动作流畅优雅的航拍视角。</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(得分：86.08%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_91_0.gif" width=500>
</div>
<div>
<h3 class="score-amount">Hunyuan </h3>
<div class="score-percentage">(得分：13.92%)</div>
<img src="https://assets.rapidata.ai/hunyuan_0091_421.gif" width=500>
</div>
</div>
</div>
<div class="container">
<div class="text-center">
<q>宇宙飞船内宇航员透过舷窗眺望地球的超写实画面，柔和环境光映衬控制面板，营造出宁静又震撼的氛围。</q>
</div>
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(得分：11.26%)</div>
<img src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_59_0.gif" width=500>
</div>
<div>
<h3 class="score-amount">Sora </h3>
<div class="score-percentage">(得分：88.74%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/sora_0059_0.gif" width=500>
</div>
</div>
</div>
</div>
# 连贯性 (Coherence)
连贯性得分用于评估生成视频的逻辑一致性以及是否存在视觉伪影 (artifacts)。标注员在看不到原始提示词的情况下，被要求回答："哪一段视频存在更多瑕疵，更像是AI生成内容？"
## 示例
<div class="vertical-container">
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(瑕疵评分：7.35%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_36_0.gif" width="500" alt="Dataset visualization">
</div>
<div>
<h3 class="score-amount">Sora </h3>
<div class="score-percentage">(瑕疵评分：92.75%)</div>
<img src="https://assets.rapidata.ai/sora_0036_0.gif" width="500" alt="Dataset visualization">
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(瑕疵评分：87.97%)</div>
<img src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_37_0.gif" width="500" alt="Dataset visualization">
</div>
<div>
<h3 class="score-amount">Ray 2 </h3>
<div class="score-percentage">(瑕疵评分：12.03%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/ray2_0037_2.gif" width="500" alt="Dataset visualization">
</div>
</div>
</div>
</div>
# 审美偏好 (Preference)
审美偏好得分用于反映参与者对视频视觉美观度的评价，与原始提示词无关。标注员被要求回答："你更倾向于哪一段视频的美学表现？"
## 示例
<div class="vertical-container">
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(得分：94.77%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_11_0.gif" width="500" alt="Dataset visualization">
</div>
<div>
<h3 class="score-amount">Pika </h3>
<div class="score-percentage">(得分：5.23%)</div>
<img src="https://assets.rapidata.ai/pika_0011_2286430682.gif" width="500" alt="Dataset visualization">
</div>
</div>
</div>
<div class="container">
<div class="image-container">
<div>
<h3 class="score-amount">Kling v2.1 Master </h3>
<div class="score-percentage">(得分：14.66%)</div>
<img src="https://assets.rapidata.ai/kling-v2.1-master-25-7-25_64_0.gif" width="500" alt="Dataset visualization">
</div>
<div>
<h3 class="score-amount">Seedance 1 Pro </h3>
<div class="score-percentage">(得分：85.34%)</div>
<img style="border: 5px solid #18c54f;" src="https://assets.rapidata.ai/seedance-1-pro-24-7-25_64_0.gif " width="500" alt="Dataset visualization">
</div>
</div>
</div>
</div>
# 关于 Rapidata
Rapidata的技术使大规模收集人类反馈的过程比以往任何时候都更快速、更便捷。访问[rapidata.ai](https://www.rapidata.ai/)了解更多关于我们如何革新AI开发中的人类反馈采集技术的信息。
# 其他数据集
我们针对主流视频生成模型开展了基准测试，测试结果可在[官方网站](https://www.rapidata.ai/leaderboard/video-models)查看。我们从连贯性/合理性、与提示词的对齐性以及风格偏好三个维度对模型进行排名。背后包含200万+条标注数据的数据集可通过以下链接获取：
- 链接至[Rich Video Annotation数据集](https://huggingface.co/datasets/Rapidata/text-2-video-Rich-Human-Feedback)
- 链接至[连贯性数据集](https://huggingface.co/datasets/Rapidata/Flux_SD3_MJ_Dalle_Human_Coherence_Dataset)
- 链接至[文本-图像对齐数据集](https://huggingface.co/datasets/Rapidata/Flux_SD3_MJ_Dalle_Human_Alignment_Dataset)
- 链接至[偏好数据集](https://huggingface.co/datasets/Rapidata/700k_Human_Preference_Dataset_FLUX_SD3_MJ_DALLE3)
提供机构：
maas
创建时间：
2025-08-01
