English
Related papers

Related papers: Expressive architectures enhance interpretability …

200 papers

Unsupervised approaches to large language model (LLM) interpretability, such as sparse autoencoders (SAEs), offer a way to decode LLM activations into interpretable and, ideally, controllable concepts. On the one hand, these approaches…

Machine Learning · Computer Science 2026-03-03 Shruti Joshi , Andrea Dittadi , Sébastien Lachapelle , Dhanya Sridhar

Time-lagged autoencoders (TAEs) have been proposed as a deep learning regression-based approach to the discovery of slow modes in dynamical systems. However, a rigorous analysis of nonlinear TAEs remains lacking. In this work, we discuss…

Machine Learning · Statistics 2019-09-04 Wei Chen , Hythem Sidky , Andrew L. Ferguson

Understanding the internal representations of large language models (LLMs) remains a central challenge for interpretability research. Sparse autoencoders (SAEs) offer a promising solution by decomposing activations into interpretable…

Machine Learning · Computer Science 2025-10-10 Yifei Yao , Mengnan Du

Operator regression provides a powerful means of constructing discretization-invariant emulators for partial-differential equations (PDEs) describing physical systems. Neural operators specifically employ deep neural networks to approximate…

Machine Learning · Computer Science 2023-04-18 Katiana Kontolati , Somdatta Goswami , George Em Karniadakis , Michael D. Shields

Understanding and modeling nonlinear dynamical systems is a fundamental challenge across science and engineering. Deep learning has shown remarkable potential for capturing complex system behavior, yet achieving models that are both…

Machine Learning · Computer Science 2026-03-06 Wei Liu , Kiran Bacsa , Loon Ching Tang , Eleni Chatzi

Partial Differential Equations (PDEs) with high dimensionality are commonly encountered in computational physics and engineering. However, finding solutions for these PDEs can be computationally expensive, making model-order reduction…

Machine Learning · Statistics 2023-03-07 Sebastian Kaltenbach , Phaedon-Stelios Koutsourelakis , Petros Koumoutsakos

A data-driven framework is proposed towards the end of predictive modeling of complex spatio-temporal dynamics, leveraging nested non-linear manifolds. Three levels of neural networks are used, with the goal of predicting the future state…

Computational Physics · Physics 2020-09-14 Jiayang Xu , Karthik Duraisamy

Low-rank recurrent neural networks (lrRNNs) are a class of models that uncover low-dimensional latent dynamics underlying neural population activity. Although their functional connectivity is low-rank, it lacks disentanglement…

Neurons and Cognition · Quantitative Biology 2025-11-19 Chengrui Li , Yunmiao Wang , Yule Wang , Weihan Li , Dieter Jaeger , Anqi Wu

The data-driven learning of solutions of partial differential equations can be based on a divide-and-conquer strategy. First, the high dimensional data is compressed to a latent space with an autoencoder; and, second, the temporal dynamics…

Machine Learning · Computer Science 2024-10-24 Elise Özalp , Luca Magri

Recurrent neural networks (RNNs) are powerful models for processing time-series data, but it remains challenging to understand how they function. Improving this understanding is of substantial interest to both the machine learning and…

Machine Learning · Computer Science 2021-11-03 Jimmy T. H. Smith , Scott W. Linderman , David Sussillo

Sparse autoencoders (SAEs) have shown promise in extracting interpretable features from complex neural networks. We present one of the first applications of SAEs to dense text embeddings from large language models, demonstrating their…

Machine Learning · Computer Science 2024-08-06 Charles O'Neill , Christine Ye , Kartheik Iyer , John F. Wu

Recurrent neural networks (RNNs) have yielded promising results for both recognizing objects in challenging conditions and modeling aspects of primate vision. However, the representational dynamics of recurrent computations remain poorly…

Computer Vision and Pattern Recognition · Computer Science 2023-10-13 Sushrut Thorat , Adrien Doerig , Tim C. Kietzmann

The advent of large-scale neural recordings has enabled new methods to discover the computational mechanisms of neural circuits by understanding the rules that govern how their state evolves over time. While these \textit{neural dynamics}…

Neurons and Cognition · Quantitative Biology 2023-09-13 Christopher Versteeg , Andrew R. Sedler , Jonathan D. McCart , Chethan Pandarinath

Sparse autoencoders (SAEs) are widely used to extract sparse, interpretable latents from transformer activations. We test whether commonly used SAE quality metrics and automatic explanation pipelines can distinguish trained transformers…

Machine Learning · Computer Science 2026-01-28 Thomas Heap , Tim Lawson , Lucy Farnik , Laurence Aitchison

Decomposing model activations into interpretable components is a key open problem in mechanistic interpretability. Sparse autoencoders (SAEs) are a popular method for decomposing the internal activations of trained transformers into sparse,…

Machine Learning · Computer Science 2024-06-26 Connor Kissane , Robert Krzyzanowski , Joseph Isaac Bloom , Arthur Conmy , Neel Nanda

The manifold assumption for high-dimensional data assumes that the data is generated by varying a set of parameters obtained from a low-dimensional latent space. Deep generative models (DGMs) are widely used to learn data representations in…

Machine Learning · Computer Science 2022-07-19 Krithika Iyer , Riddhish Bhalodia , Shireen Elhabian

Sparse autoencoders (SAEs) are a mechanistic interpretability technique that have been used to provide insight into learned concepts within large protein language models. Here, we employ TopK and Ordered SAEs to investigate autoregressive…

Variational Autoencoders (VAEs) have become a popular approach for dimensionality reduction. However, despite their ability to identify latent low-dimensional structures embedded within high-dimensional data, these latent representations…

Machine Learning · Statistics 2020-08-27 Kaspar Märtens , Christopher Yau

Data-driven reduced-order models based on autoencoders generally lack interpretability compared to classical methods such as the proper orthogonal decomposition. More interpretability can be gained by disentangling the latent variables and…

Machine Learning · Computer Science 2025-02-21 Henning Schwarz , Pyei Phyo Lin , Jens-Peter M. Zemke , Thomas Rung

Sequence-to-sequence (Seq2seq) models have played an important role in the recent success of various natural language processing methods, such as machine translation, text summarization, and speech recognition. However, current Seq2seq…

Computation and Language · Computer Science 2018-06-05 Myeongjun Jang , Seungwan Seo , Pilsung Kang
‹ Prev 1 3 4 5 6 7 10 Next ›