paper-with-me

홈 › Papers

Tree SAE: Learning Hierarchical Feature Structures in Sparse Autoencoders

2026-05-08 · Tue M. Cao, Hoang X. Nhat, Raed Alharbi, Phi Le Nguyen, My T. Thai arxiv

Learning hierarchical features in Sparse Autoencoders (SAEs) is essential for capturing the structured nature of real-world data and mitigating issues like feature absorption or splitting. Existing works attempt to identify hierarchical relationships within independent feature sets by relying on activation coverage, the assumption that child feature should only activate when its parent feature activates. However, we demonstrate that this condition alone is insufficient; that is, it often produces false positives where parent and child concepts are semantically unrelated. To address this, we introduce a novel reconstruction condition that enforces a deeper functional link between hierarchical levels. By combining both activation and reconstruction constraints, we propose the Tree SAE, a model designed to learn hierarchical structures directly from within the feature set. Our results demonstrate that Tree SAEs significantly surpass the existing SAEs at learning hierarchical pairs while maintaining competitive performance to the state-of-the-art on several key benchmarks. Finally, we demonstrate the practical utility of our Tree SAE in mapping the geometry of child feature subspaces and uncovering the complex hierarchical concept structures encoded within large language models.

📄 PDF Abstract BibTeX arXiv:2605.07922

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

From Atoms to Trees: Building a Structured Feature Forest with Hierarchical Sparse Autoencoders

2026-02-12 · Yifan Luo, Yang Zhan, Jiedong Jiang, Tianyang Liu 외 arxiv

Sparse autoencoders (SAEs) have proven effective for extracting monosemantic features from large language models (LLMs), yet these features are typically identified in isolation. However, broad evidence suggests that LLM…

Winner-Take-All Autoencoders

2014-09-09 · NeurIPS 2015 12 · Alireza Makhzani, Brendan Frey

In this paper, we propose a winner-take-all method for learning hierarchical sparse representations in an unsupervised fashion. We first introduce fully-connected winner-take-all autoencoders which use mini-batch statist…

All

Coupled Hierarchical Structure Learning using Tree-Wasserstein Distance

2025-01-07 · Ya-Wei Eileen Lin, Ronald R. Coifman, Gal Mishne, Ronen Talmon

In many applications, both data samples and features have underlying hierarchical structures. However, existing methods for learning these latent structures typically focus on either samples or features, ignoring possibl…

Link PredictionNode Classification

Tree Variational Autoencoders

2023-09-21 · NeurIPS 2023 11

We propose Tree Variational Autoencoder (TreeVAE), a new generative hierarchical clustering model that learns a flexible tree-based posterior distribution over latent variables. TreeVAE hierarchically divides samples a…

Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models

2025-04-03 · Mateusz Pach, Shyamgopal Karthik, Quentin Bouniot, Serge Belongie 외

Sparse Autoencoders (SAEs) have recently been shown to enhance interpretability and steerability in Large Language Models (LLMs). In this work, we extend the application of SAEs to Vision-Language Models (VLMs), such as …