English
Related papers

Related papers: Mechanistic Data Attribution: Tracing the Training…

200 papers

Training data attribution (TDA) methods aim to identify which training examples influence a model's predictions on specific test data most. By quantifying these influences, TDA supports critical applications such as data debugging,…

Machine Learning · Computer Science 2025-05-30 Xingyuan Pan , Chenlu Ye , Joseph Melkonian , Jiaqi W. Ma , Tong Zhang

Training data attribution (TDA) plays a critical role in understanding the influence of individual training data points on model predictions. Gradient-based TDA methods, popularized by \textit{influence function} for their superior…

Machine Learning · Computer Science 2025-09-17 Shiyuan Zhang , Junwei Deng , Juhan Bae , Jiaqi Ma

Statistical learning relies upon data sampled from a distribution, and we usually do not care what actually generated it in the first place. From the point of view of causal modeling, the structure of each distribution is induced by…

Machine Learning · Computer Science 2018-09-11 Giambattista Parascandolo , Niki Kilbertus , Mateo Rojas-Carulla , Bernhard Schölkopf

Mechanical ventilation is a critical life support intervention that delivers controlled air and oxygen to a patient's lungs, assisting or replacing spontaneous breathing. While several data-driven approaches have been proposed to optimize…

Machine Learning · Computer Science 2025-01-10 Joo Seung Lee , Malini Mahendra , Anil Aswani

The discovery of causal relationships from high-dimensional data is a major open problem in bioinformatics. Machine learning and feature attribution models have shown great promise in this context but lack causal interpretation. Here, we…

Machine Learning · Computer Science 2023-04-26 Payam Dibaeinia , Saurabh Sinha

Mechanistic interpretability has identified small sets of attention heads that implement specific behaviours in transformer language models, but recovering these circuits typically requires a bespoke analytical pipeline for each new task.…

Machine Learning · Computer Science 2026-05-27 Barsat Khadka

The goal of data attribution is to trace the model's predictions through the learning algorithm and back to its training data. thereby identifying the most influential training samples and understanding how the model's behavior leads to…

Machine Learning · Computer Science 2025-08-12 Hongbo Zhu , Angelo Cangelosi

Large Language Models (LLMs) have experienced a rapid rise in AI, changing a wide range of applications with their advanced capabilities. As these models become increasingly integral to decision-making, the need for thorough…

Machine Learning · Computer Science 2024-01-09 Jatin Nainani

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

Current training data attribution (TDA) methods treat the influence one sample has on another as static, but neural networks learn in distinct stages that exhibit changing patterns of influence. In this work, we introduce a framework for…

Machine Learning · Computer Science 2025-10-15 Jin Hwa Lee , Matthew Smith , Maxwell Adam , Jesse Hoogland

Mechanistic interpretability has transformed the analysis of transformer circuits by decomposing model behavior into competing algorithms, identifying phase transitions during training, and deriving closed-form predictions for when and why…

Machine Learning · Computer Science 2026-03-19 Alma Lago

Multi-Agent Deep Reinforcement Learning (MADRL) was proven efficient in solving complex problems in robotics or games, yet most of the trained models are hard to interpret. While learning intrinsically interpretable models remains a…

Artificial Intelligence · Computer Science 2025-02-04 Yoann Poupart , Aurélie Beynier , Nicolas Maudet

Attribution theory explains how individuals interpret and attribute others' behavior in a social context by employing personal (dispositional) and impersonal (situational) causality. Large Language Models (LLMs), trained on human-generated…

Computation and Language · Computer Science 2026-03-31 Hossein Salemi , Jitin Krishnan , Hemant Purohit

Large language models (LLMs) are often portrayed as merely imitating linguistic patterns without genuine understanding. We argue that recent findings in mechanistic interpretability (MI), the emerging field probing the inner workings of…

Computation and Language · Computer Science 2026-02-26 Pierre Beckmann , Matthieu Queloz

Data attribution seeks to trace model behavior back to the training examples that shaped it, enabling debugging, auditing, and data valuation at scale. Classical influence-function methods offer a principled foundation but remain…

Machine Learning · Computer Science 2025-11-26 Sibo Ma , Julian Nyarko

Training data attribution (TDA) techniques find influential training data for the model's prediction on the test data of interest. They approximate the impact of down- or up-weighting a particular training sample. While conceptually useful,…

Machine Learning · Computer Science 2023-11-01 Elisa Nguyen , Minjoon Seo , Seong Joon Oh

Causal discovery traditionally relies on statistical methods applied to observational data, often requiring large datasets and assumptions about underlying causal structures. Recent advancements in Large Language Models (LLMs) have…

Machine Learning · Computer Science 2025-04-16 Yuni Susanti , Michael Färber

Explainable AI (XAI) aims to provide insight into opaque model reasoning to humans and as such is an interdisciplinary field by nature. In this paper, we interviewed 10 practitioners to understand the possible usability of training data…

Human-Computer Interaction · Computer Science 2023-11-23 Elisa Nguyen , Evgenii Kortukov , Jean Y. Song , Seong Joon Oh

When comparing inductive logic programming (ILP) and attribute-value learning techniques, there is a trade-off between expressive power and efficiency. Inductive logic programming techniques are typically more expressive but also less…

Machine Learning · Computer Science 2007-05-23 Hendrik Blockeel , Luc De Raedt , Nico Jacobs , Bart Demoen

Large language models (LLMs) have achieved remarkable capabilities across diverse tasks, yet their internal decision-making processes remain largely opaque. Mechanistic interpretability (i.e., the systematic study of how neural networks…

Computation and Language · Computer Science 2026-02-13 Usman Naseem