ClusterLLM:作为文本聚类引导的大语言模型
ClusterLLM: Large Language Models as a Guide for Text Clustering
AI总结:
本研究提出ClusterLLM文本聚类框架,借助ChatGPT反馈,通过难三元组问题微调嵌入器、成对问题调整聚类粒度,在14个数据集上以低成本提升了聚类质量。
摘要:
我们提出了ClusterLLM,这是一种新颖的文本聚类框架,它利用经过指令微调的大语言模型(如ChatGPT)的反馈来实现聚类。与基于“小型”嵌入器构建的传统无监督方法相比,ClusterLLM具有两个显著优势:(1)即使无法获取LLM的嵌入向量,它也能利用LLM的涌现能力;(2)它能通过文本指令和/或少量标注数据理解用户的聚类偏好。首先,我们通过构造难三元组问题<A是否比C更对应B>来向ChatGPT询问聚类视角的见解,其中A、B、C是根据小型嵌入器判定属于不同簇的相似数据点。我们通过实验证明,该策略既能有效微调小型嵌入器,查询ChatGPT的成本也很低。其次,我们通过精心设计的成对问题<A和B是否属于同一类别>向ChatGPT寻求聚类粒度的帮助,并从与ChatGPT答案最一致的簇层次结构中调整粒度。在14个数据集上的大量实验表明,ClusterLLM持续提升了聚类质量,每个数据集的平均成本约为0.6美元。代码将发布于https://github.com/zhang-yu-wei/ClusterLLM。
英文摘要:
We introduce ClusterLLM, a novel text clustering framework that leverages feedback from an instruction-tuned large language model, such as ChatGPT. Compared with traditional unsupervised methods that builds upon "small" embedders, ClusterLLM exhibits two intriguing advantages: (1) it enjoys the emergent capability of LLM even if its embeddings are inaccessible; and (2) it understands the user's preference on clustering through textual instruction and/or a few annotated data. First, we prompt ChatGPT for insights on clustering perspective by constructing hard triplet questions <does A better correspond to B than C>, where A, B and C are similar data points that belong to different clusters according to small embedder. We empirically show that this strategy is both effective for fine-tuning small embedder and cost-efficient to query ChatGPT. Second, we prompt ChatGPT for helps on clustering granularity by carefully designed pairwise questions <do A and B belong to the same category>, and tune the granularity from cluster hierarchies that is the most consistent with the ChatGPT answers. Extensive experiments on 14 datasets show that ClusterLLM consistently improves clustering quality, at an average cost of ~$0.6 per dataset. The code will be available at https://github.com/zhang-yu-wei/ClusterLLM.
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。