用大模型辅助主题模型，后处理比初始化更有效。
Semantic-Augmented Latent Topic Modeling with LLM-in-the-Loop
- 大模型用于主题模型的初始化与修正两个阶段
- 后处理使主题连贯性提升5.86%
- 实验表明大模型并非总是优于传统方法
隐含狄利克雷分配（LDA）是一种广泛使用的生成概率模型，用于挖掘文档集合中的抽象主题。本文探索将大语言模型（LLM）融入主题模型的两个关键阶段：初始化与后修正。由于LDA对初始化质量高度敏感，我们通过大模型引导的主题聚类来初始化吉布斯采样算法，进行了大量实验。结果显示，该初始化策略虽改善了早期迭代，但对收敛无影响，且性能劣于基线。相比之下，大模型驱动的后修正在主题连贯性评估中实现了5.86%的显著提升。这些结果揭示了'大模型在环'方法的实际价值，并挑战了大模型始终是文本挖掘更优选择的普遍认知。
原文摘要 · Abstract (English)
Latent Dirichlet Allocation (LDA) is a prominent generative probabilistic model used for uncovering abstract topics within document collections. In this paper, we explore the effectiveness of augmenting topic models with Large Language Models (LLMs) through integration into two key phases: Initialization and Post-Correction. Since the LDA is highly dependent on the quality of its initialization, we conduct extensive experiments on the LLM-guided topic clustering for initializing the Gibbs sampling algorithm. Interestingly, the experimental results reveal that while the proposed initialization strategy improves the early iterations of LDA, it has no effect on the convergence and yields the worst performance compared to the baselines. The LLM-enabled post-correction, on the other hand, achieved a promising improvement of 5.86% in the coherence evaluation. These results highlight the practical benefits of the LLM-in-the-loop approach and challenge the belief that LLMs are always the superior text mining alternative.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。