将稀疏自编码器视为主题模型，实现跨模态主题分析
Sparse Autoencoders are Topic Models
- 将SAE视为连续主题模型，从贝叶斯推断角度推导其目标函数
- SAE-TM框架生成更连贯且多样化的主题，无需重新训练
- 适用于文本与图像数据的主题结构分析，可追踪主题演化
稀疏自编码器（SAEs）常用于嵌入空间分析，但其作用和实际价值存在争议。本文提出新视角：SAE可自然理解为一种主题模型。受潜在狄利克雷分配（LDA）启发，我们构建连续主题模型（CTM），并推导出SAE目标函数为其最大后验估计。该视角表明SAE特征是主题成分而非可操控方向。为验证理论，我们提出SAE-TM框架：(1) 训练SAE以学习可复用的主题原子；(2) 将其解释为下游数据上的词分布；(3) 无需重新训练即可合并成任意数量主题。在文本与图像数据集上，SAE-TM生成的主题比强基线更连贯且保持多样性。最后，我们分析了图像数据的主题结构，并追踪日本版画中主题随时间的变化。本工作将SAE定位为跨模态大规模主题分析的有效工具。代码已开源。
原文摘要 · Abstract (English)
Sparse autoencoders (SAEs) are used to analyze embeddings, but their role and practical value are debated. We propose a new perspective on SAEs by demonstrating that they can be naturally understood as topic models. We propose a continuous topic model (CTM) inspired by Latent Dirichlet Allocation (LDA) for embedding spaces and derive the SAE objective as a maximum a posteriori estimator under this model. This view implies SAE features are thematic components rather than steerable directions. To confirm our theoretical findings, we introduce SAE-TM, a topic modeling framework that: (1) trains an SAE to learn reusable topic atoms, (2) interprets them as word distributions on downstream data, and (3) merges them into any number of topics without retraining. SAE-TM yields more coherent topics than strong baselines on text and image datasets while maintaining diversity. Finally, we analyze thematic structure in image datasets and trace topic changes over time in Japanese woodblock prints. Our work positions SAEs as effective tools for large-scale thematic analysis across modalities. Code is available at https://github.com/ExplainableML/SAE-TM .
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。