利用机器学习和大语言模型实现法律发现中的自动化图像聚类与描述
Leveraging Machine Learning and Large Language Models for Automated Image Clustering and Description in Legal Discovery
AI总结：
本文提出利用机器学习和大语言模型实现法律发现中自动化图像聚类与描述，通过比较不同采样策略、提示技术和生成方法，提升大规模图像数据处理的效率和准确性。
摘要：
数字图像的快速增加在法律发现、数字档案和内容管理中带来了重大挑战。企业及法律团队必须在严格的时间压力下对大量图像进行组织、分析和提取有意义的洞察，手动审查不切实际且成本高昂。这些需求加剧了对能够高效组织和描述大规模图像数据集的自动化方法的兴趣。本文通过整合图像聚类、图像描述和大语言模型（LLMs）进行系统研究，探讨自动化聚类描述生成。我们使用K-means聚类将图像分成20个视觉一致的聚类，并使用Azure AI Vision API生成基础描述。然后我们评估了聚类描述过程的三个关键维度：（1）图像采样策略，比较随机、基于质心、分层、混合和基于密度的采样方法与使用所有聚类图像；（2）提示技术，比较标准提示与思维链提示；（3）描述生成方法，比较基于LLM的生成与传统TF-IDF和基于模板的方法。我们使用语义相似性和覆盖度指标评估描述质量。结果显示，使用每个聚类20张图像的策略在性能上与全面包含相当，但显著降低了计算成本，只有分层采样显示出轻微的退化。基于LLM的方法在TF-IDF基线中表现一致，标准提示在该任务中优于思维链提示。这些发现为部署可扩展、准确的聚类描述系统提供了实用指导，这些系统支持法律发现和其他需要自动化组织大规模图像集合的领域中的高吞吐量工作流程。
英文摘要：
The rapid increase in digital image creation and retention presents substantial challenges during legal discovery, digital archive, and content management. Corporations and legal teams must organize, analyze, and extract meaningful insights from large image collections under strict time pressures, making manual review impractical and costly. These demands have intensified interest in automated methods that can efficiently organize and describe large-scale image datasets. This paper presents a systematic investigation of automated cluster description generation through the integration of image clustering, image captioning, and large language models (LLMs). We apply K-means clustering to group images into 20 visually coherent clusters and generate base captions using the Azure AI Vision API. We then evaluate three critical dimensions of the cluster description process: (1) image sampling strategies, comparing random, centroid-based, stratified, hybrid, and density-based sampling against using all cluster images; (2) prompting techniques, contrasting standard prompting with chain-of-thought prompting; and (3) description generation methods, comparing LLM-based generation with traditional TF-IDF and template-based approaches. We assess description quality using semantic similarity and coverage metrics. Results show that strategic sampling with 20 images per cluster performs comparably to exhaustive inclusion while significantly reducing computational cost, with only stratified sampling showing modest degradation. LLM-based methods consistently outperform TF-IDF baselines, and standard prompts outperform chain-of-thought prompts for this task. These findings provide practical guidance for deploying scalable, accurate cluster description systems that support high-volume workflows in legal discovery and other domains requiring automated organization of large image collections.
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。
