English
Related papers

Related papers: The Geometry of Concepts: Sparse Autoencoder Featu…

200 papers

Given a pool of observations selected from a sensor stream, input data can be robustly represented, via a multiscale process, in terms of invariant concepts, and themes. Applying this to episodic natural language data, one may obtain a…

Artificial Intelligence · Computer Science 2020-10-19 Mark Burgess

Sparse autoencoders (SAEs) have gained a lot of attention as a promising tool to improve the interpretability of large language models (LLMs) by mapping the complex superposition of polysemantic neurons into monosemantic features and…

Computation and Language · Computer Science 2025-02-19 Gouki Minegishi , Hiroki Furuta , Yusuke Iwasawa , Yutaka Matsuo

We consider a novel clustering task in which clusters can have compositional relationships, e.g., one cluster contains images of rectangles, one contains images of circles, and a third (compositional) cluster contains images with both…

Machine Learning · Computer Science 2023-08-04 Zeqian Li , Xinlu He , Jacob Whitehill

Sparse Autoencoders (SAEs) have emerged as a predominant tool in mechanistic interpretability, aiming to identify interpretable monosemantic features. However, how does sparse encoding organize the representations of activation vector from…

Machine Learning · Computer Science 2025-05-29 Wenjie Sun , Bingzhe Wu , Zhile Yang , Chengke Wu

Autoencoders offer a general way of learning low-dimensional, non-linear representations from data without labels. This is achieved without making any particular assumptions about the data type or other domain knowledge. The generality and…

Machine Learning · Computer Science 2025-05-27 Collin Leiber , Lukas Miklautz , Claudia Plant , Christian Böhm

Previous studies primarily utilize MLP neurons as units of analysis for understanding the mechanisms of factual knowledge in Language Models (LMs); however, neurons suffer from polysemanticity, leading to limited knowledge expression and…

Computation and Language · Computer Science 2025-02-28 Yuheng Chen , Pengfei Cao , Kang Liu , Jun Zhao

State-of-the-art approaches for clustering high-dimensional data utilize deep auto-encoder architectures. Many of these networks require a large number of parameters and suffer from a lack of interpretability, due to the black-box nature of…

Machine Learning · Computer Science 2022-02-28 Alexander Lin , Andrew H. Song , Demba Ba

Large Language Models (LLMs) have achieved strong complex reasoning capabilities through Chain-of-Thought (CoT) reasoning. However, their reasoning patterns remain too complicated to analyze. While Sparse Autoencoders (SAEs) have emerged as…

Machine Learning · Computer Science 2026-03-04 Xuan Yang , Jiayu Liu , Yuhang Lai , Hao Xu , Zhenya Huang , Ning Miao

In this paper, we present a novel neuroevolutionary method to identify the architecture and hyperparameters of convolutional autoencoders. Remarkably, we used a hypervolume indicator in the context of neural architecture search for…

Neural and Evolutionary Computing · Computer Science 2021-06-23 Daniel Dimanov , Emili Balaguer-Ballester , Colin Singleton , Shahin Rostami

Face clustering can provide pseudo-labels to the massive unlabeled face data and improve the performance of different face recognition models. The existing clustering methods generally aggregate the features within subgraphs that are often…

Computer Vision and Pattern Recognition · Computer Science 2023-04-24 Yuan Cao , Di Jiang , Guanqun Hou , Fan Deng , Xinjia Chen , Qiang Yang

We study the Dictionary Learning (aka Sparse Coding) problem of obtaining a sparse representation of data points, by learning \emph{dictionary vectors} upon which the data points can be written as sparse linear combinations. We view this…

Machine Learning · Computer Science 2015-03-09 Meera Sitharam , Mohamad Tarifi , Menghan Wang

Sparse Autoencoders (SAEs) have emerged as a powerful framework for machine learning interpretability, enabling the unsupervised decomposition of model representations into a dictionary of abstract, human-interpretable concepts. However, we…

Computer Vision and Pattern Recognition · Computer Science 2025-05-27 Thomas Fel , Ekdeep Singh Lubana , Jacob S. Prince , Matthew Kowal , Victor Boutin , Isabel Papadimitriou , Binxu Wang , Martin Wattenberg , Demba Ba , Talia Konkle

This paper aims to improve the feature learning in Convolutional Networks (Convnet) by capturing the structure of objects. A new sparsity function is imposed on the extracted featuremap to capture the structure and shape of the learned…

Machine Learning · Computer Science 2017-01-03 Ehsan Hosseini-Asl

Word embeddings are rich word representations, which in combination with deep neural networks, lead to large performance gains for many NLP tasks. However, word embeddings are represented by dense, real-valued vectors and they are therefore…

Computation and Language · Computer Science 2019-12-24 Andreas Hanselowski , Iryna Gurevych

Sparse autoencoders (SAEs) have emerged as a promising approach in language model interpretability, offering unsupervised extraction of sparse features. For interpretability methods to succeed, they must identify abstract features across…

Machine Learning · Computer Science 2025-09-08 Lovis Heindrich , Philip Torr , Fazl Barez , Veronika Thost

We investigate the 3-dimensional Edwards-Anderson spin glass model at low temperature on simple cubic lattices of sizes up to L=12. Our findings show a strong continuity among T>0 physical features and those found previously at T=0, leading…

Disordered Systems and Neural Networks · Physics 2009-11-07 E. Marinari , O. C. Martin , F. Zuliani

Networks with underlying metric spaces attract increasing research attention in network science, statistical physics, applied mathematics, computer science, sociology, and other fields. This attention is further amplified by the current…

Physics and Society · Physics 2020-04-22 Marian Boguna , Dmitri Krioukov , Pedro Almagro , M. Angeles Serrano

Autoencoders have been widely used as a nonlinear tool for data dimensionality reduction. While autoencoders don't utilize the label information, Centroid-Encoders (CE)\cite{ghosh2022supervised} use the class label in their learning…

Machine Learning · Computer Science 2022-06-30 Tomojit Ghosh , Michael Kirby

Sparse auto-encoders (SAEs) have re-emerged as a prominent method for mechanistic interpretability, yet they face two significant challenges: the non-smoothness of the $L_1$ penalty, which hinders reconstruction and scalability, and a lack…

Artificial Intelligence · Computer Science 2026-05-19 Ouns El Harzli , Hugo Wallner , Yoonsoo Nam , Haixuan Xavier Tao

Integrating compositional and symbolic properties into current distributional semantic spaces can enhance the interpretability, controllability, compositionality, and generalisation capabilities of Transformer-based auto-regressive language…

Computation and Language · Computer Science 2026-04-16 Yingji Zhang , Danilo S. Carvalho , André Freitas
‹ Prev 1 3 4 5 6 7 10 Next ›