English
Related papers

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

200 papers

Transformer-based LLMs demonstrate strong performance on graph reasoning tasks, yet their internal mechanisms remain underexplored. To uncover these reasoning process mechanisms in a fundamental and unified view, we set the basic…

Machine Learning · Computer Science 2025-09-25 Xinnan Dai , Chung-Hsiang Lo , Kai Guo , Shenglai Zeng , Dongsheng Luo , Jiliang Tang

The usage of transformers has grown from learning about language semantics to forming meaningful visiolinguistic representations. These architectures are often over-parametrized, requiring large amounts of computation. In this work, we…

Computation and Language · Computer Science 2020-07-09 Prajjwal Bhargava

Deep learning models such as the Transformer are often constructed by heuristics and experience. To provide a complementary foundation, in this work we study the following problem: Is it possible to find an energy function underlying the…

Machine Learning · Computer Science 2023-02-28 Yongyi Yang , Zengfeng Huang , David Wipf

Self-attention mechanism is the key of the Transformer but often criticized for its computation demands. Previous token pruning works motivate their methods from the view of computation redundancy but still need to load the full network and…

Computer Vision and Pattern Recognition · Computer Science 2024-04-09 Sihao Lin , Pumeng Lyu , Dongrui Liu , Tao Tang , Xiaodan Liang , Andy Song , Xiaojun Chang

Large Language Models such as GPTs (Generative Pre-trained Transformers) exhibit remarkable capabilities across a broad spectrum of applications. Nevertheless, due to their intrinsic complexity, these models present substantial challenges…

Machine Learning · Computer Science 2024-10-17 Ashkan Golgoon , Khashayar Filom , Arjun Ravi Kannan

Through considerable effort and intuition, several recent works have reverse-engineered nontrivial behaviors of transformer models. This paper systematizes the mechanistic interpretability process they followed. First, researchers choose a…

Machine Learning · Computer Science 2023-10-31 Arthur Conmy , Augustine N. Mavor-Parker , Aengus Lynch , Stefan Heimersheim , Adrià Garriga-Alonso

Mechanistic interpretability seeks to understand the neural mechanisms that enable specific behaviors in Large Language Models (LLMs) by leveraging causality-based methods. While these approaches have identified neural circuits that copy…

Computation and Language · Computer Science 2023-08-29 Vedant Palit , Rohan Pandey , Aryaman Arora , Paul Pu Liang

Mechanistic interpretability produces circuit-level causal analyses of neural network behaviour, but discovered circuits often remain isolated experimental artefacts: there is no shared formal representation for what circuits compute, how…

Machine Learning · Computer Science 2026-05-21 Nura Aljaafari , Danilo S. Carvalho , Andre Freitas

The process by which Large Language Models (LLMs) acquire complex capabilities during training remains a key open question in mechanistic interpretability. This project investigates whether these learning dynamics can be characterized…

Machine Learning · Computer Science 2025-09-22 Elisabetta Rocchetti

Transformers were initially introduced for natural language processing (NLP) tasks, but fast they were adopted by most deep learning fields, including computer vision. They measure the relationships between pairs of input tokens (words in…

Computer Vision and Pattern Recognition · Computer Science 2023-03-22 Robin Courant , Maika Edberg , Nicolas Dufour , Vicky Kalogeiton

The transformer architecture is widely used in machine learning models and consists of two alternating sublayers: attention heads and MLPs. We prove that an MLP neuron can be implemented by a masked attention head with internal dimension 1…

Machine Learning · Computer Science 2023-09-18 Robert Huben , Valerie Morris

The Transformer architecture has achieved tremendous success in natural language processing, computer vision, and scientific computing through its self-attention mechanism. However, its core components-positional encoding and attention…

Machine Learning · Computer Science 2025-11-13 Xianshuai Shi , Jianfeng Zhu , Leibo Liu

Capturing interpretable variations has long been one of the goals in disentanglement learning. However, unlike the independence assumption, interpretability has rarely been exploited to encourage disentanglement in the unsupervised setting.…

Computer Vision and Pattern Recognition · Computer Science 2021-04-13 Xinqi Zhu , Chang Xu , Dacheng Tao

Attention layers are widely used in natural language processing (NLP) and are beginning to influence computer vision architectures. Training very large transformer models allowed significant improvement in both fields, but once trained,…

Machine Learning · Computer Science 2021-05-21 Jean-Baptiste Cordonnier , Andreas Loukas , Martin Jaggi

With the continue development of Convolutional Neural Networks (CNNs), there is a growing concern regarding representations that they encode internally. Analyzing these internal representations is referred to as model interpretation. While…

Computer Vision and Pattern Recognition · Computer Science 2023-05-18 Hamed Behzadi-Khormouji , José Oramas

We propose a method to create document representations that reflect their internal structure. We modify Tree-LSTMs to hierarchically merge basic elements such as words and sentences into blocks of increasing complexity. Our Structure…

Computation and Language · Computer Science 2019-10-08 Khalil Mrini , Claudiu Musat , Michael Baeriswyl , Martin Jaggi

We introduce refined variants of the Local Learning Coefficient (LLC), a measure of model complexity grounded in singular learning theory, to study the development of internal structure in transformer language models during training. By…

Machine Learning · Computer Science 2024-10-07 George Wang , Jesse Hoogland , Stan van Wingerden , Zach Furman , Daniel Murfet

Multi-graph learning is crucial for extracting meaningful signals from collections of heterogeneous graphs. However, effectively integrating information across graphs with differing topologies, scales, and semantics, often in the absence of…

Machine Learning · Computer Science 2026-02-02 Zahra Moslemi , Ziyi Liang , Norbert Fortin , Babak Shahbaba

Mechanistic interpretability aims to explain what a neural network has learned at a nuts-and-bolts level. What are the fundamental primitives of neural network representations? Previous mechanistic descriptions have used individual neurons…

Circuit representations are becoming the lingua franca to express and reason about tractable generative and discriminative models. In this paper, we show how complex inference scenarios for these models that commonly arise in machine…

Machine Learning · Statistics 2021-02-12 Antonio Vergari , YooJung Choi , Anji Liu , Stefano Teso , Guy Van den Broeck