English
Related papers

Related papers: Beyond Components: Singular Vector-Based Interpret…

200 papers

Recent work in mechanistic interpretability has shown that behaviors in language models can be successfully reverse-engineered through circuit analysis. A common criticism, however, is that each circuit is task-specific, and thus such…

Computation and Language · Computer Science 2024-05-07 Jack Merullo , Carsten Eickhoff , Ellie Pavlick

Pretrained embeddings based on the Transformer architecture have taken the NLP community by storm. We show that they can mathematically be reframed as a sum of vector factors and showcase how to use this reframing to study the impact of…

Computation and Language · Computer Science 2022-06-09 Timothee Mickus , Denis Paperno , Mathieu Constant

Mechanistic interpretability aims to understand how models store representations by breaking down neural networks into interpretable units. However, the occurrence of polysemantic neurons, or neurons that respond to multiple unrelated…

Computer Vision and Pattern Recognition · Computer Science 2023-04-20 Laura O'Mahony , Vincent Andrearczyk , Henning Muller , Mara Graziani

Medical Image Computing (MIC) is a broad research topic covering both pixel-wise (e.g., segmentation, registration) and image-wise (e.g., classification, regression) vision tasks. Effective analysis demands models that capture both global…

Computer Vision and Pattern Recognition · Computer Science 2025-09-16 Mingyuan Meng

While concept-based interpretability methods have traditionally focused on local explanations of neural network predictions, we propose a novel framework and interactive tool that extends these methods into the domain of mechanistic…

Machine Learning · Computer Science 2025-07-09 Sofiia Chorna , Kateryna Tarelkina , Eloïse Berthier , Gianni Franchi

Graphical tensor notation is a simple way of denoting linear operations on tensors, originating from physics. Modern deep learning consists almost entirely of operations on or between tensors, so easily understanding tensor operations is…

Machine Learning · Computer Science 2024-02-06 Jordan K. Taylor

When trained on language data, do transformers learn some arbitrary computation that utilizes the full capacity of the architecture or do they learn a simpler, tree-like computation, hypothesized to underlie compositional meaning systems…

Computation and Language · Computer Science 2022-11-07 Shikhar Murty , Pratyusha Sharma , Jacob Andreas , Christopher D. Manning

Transformer becomes the state-of-the-art translation model, while it is not well studied how each intermediate component contributes to the model performance, which poses significant challenges for designing optimal architectures. In this…

Computation and Language · Computer Science 2020-11-10 Wenxuan Wang , Zhaopeng Tu

Understanding the internal mechanisms of large language models (LLMs) is integral to enhancing their reliability, interpretability, and inference processes. We present Constituent-Aware Pooling (CAP), a methodology designed to analyse how…

Computation and Language · Computer Science 2025-05-21 Nura Aljaafari , Danilo S. Carvalho , André Freitas

While Transformer-based pre-trained language models and their variants exhibit strong semantic representation capabilities, the question of comprehending the information gain derived from the additional components of PLMs remains an open…

Computation and Language · Computer Science 2024-01-04 Li Zhou , Wenyu Chen , Yong Cao , Dingyi Zeng , Wanlong Liu , Hong Qu

Pretrained transformer-based Language Models (LMs) are well-known for their ability to achieve significant improvement on NLP tasks, but their black-box nature, which leads to a lack of interpretability, has been a major concern. My…

Computation and Language · Computer Science 2024-12-06 Ximing Wen

We model the process of human full interpretation of object images, namely the ability to identify and localize all semantic features and parts that are recognized by human observers. The task is approached by dividing the interpretation of…

Computer Vision and Pattern Recognition · Computer Science 2018-02-02 Guy Ben-Yosef , Liav Assif , Shimon Ullman

The transformer is a neural network component that can be used to learn useful representations of sequences or sets of data-points. The transformer has driven recent advances in natural language processing, computer vision, and…

Machine Learning · Computer Science 2026-01-21 Richard E. Turner

While transformers have proven enormously successful in a range of tasks, their fundamental properties as models of computation are not well understood. This paper contributes to the study of the expressive capacity of transformers,…

Machine Learning · Computer Science 2025-03-31 Lena Strobl , Dana Angluin , Robert Frank

Transformers have attained outstanding performance across various modalities, owing to their simple but powerful scaled-dot-product (SDP) attention mechanisms. Researchers have attempted to migrate Transformers to graph learning, but most…

Machine Learning · Computer Science 2026-01-30 Liheng Ma , Soumyasundar Pal , Yingxue Zhang , Philip H. S. Torr , Mark Coates

As Transformers have become state-of-the-art models for natural language processing (NLP) tasks, the need to understand and explain their predictions is increasingly apparent. Especially in unsupervised applications, such as information…

Computation and Language · Computer Science 2024-05-13 Alexandros Vasileiou , Oliver Eberle

Attention mechanisms have improved the performance of NLP tasks while allowing models to remain explainable. Self-attention is currently widely used, however interpretability is difficult due to the numerous attention distributions. Recent…

Computation and Language · Computer Science 2020-10-30 Khalil Mrini , Franck Dernoncourt , Quan Tran , Trung Bui , Walter Chang , Ndapa Nakashole

In this paper we delve deep in the Transformer architecture by investigating two of its core components: self-attention and contextual embeddings. In particular, we study the identifiability of attention weights and token embeddings, and…

Computation and Language · Computer Science 2020-02-10 Gino Brunner , Yang Liu , Damián Pascual , Oliver Richter , Massimiliano Ciaramita , Roger Wattenhofer

Attention mechanisms form a core component of several successful deep learning architectures, and are based on one key idea: ''The output depends only on a small (but unknown) segment of the input.'' In several practical applications like…

Machine Learning · Computer Science 2023-05-16 Lakshmi Narayan Pandey , Rahul Vashisht , Harish G. Ramaswamy

Linear probes and sparse autoencoders consistently recover meaningful structure from transformer representations -- yet why should such simple methods succeed in deep, nonlinear systems? We show this is not merely an empirical regularity…

Machine Learning · Computer Science 2026-02-11 Andres Saurez , Yousung Lee , Dongsoo Har