Integrated Influence: Data Attribution with Baseline

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Linxiao, Gu, Xinyu, Sun, Liang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916885039153152
author Yang, Linxiao
Gu, Xinyu
Sun, Liang
author_facet Yang, Linxiao
Gu, Xinyu
Sun, Liang
contents As an effective approach to quantify how training samples influence test sample, data attribution is crucial for understanding data and model and further enhance the transparency of machine learning models. We find that prevailing data attribution methods based on leave-one-out (LOO) strategy suffer from the local-based explanation, as these LOO-based methods only perturb a single training sample, and overlook the collective influence in the training set. On the other hand, the lack of baseline in many data attribution methods reduces the flexibility of the explanation, e.g., failing to provide counterfactual explanations. In this paper, we propose Integrated Influence, a novel data attribution method that incorporates a baseline approach. Our method defines a baseline dataset, follows a data degeneration process to transition the current dataset to the baseline, and accumulates the influence of each sample throughout this process. We provide a solid theoretical framework for our method, and further demonstrate that popular methods, such as influence functions, can be viewed as special cases of our approach. Experimental results show that Integrated Influence generates more reliable data attributions compared to existing methods in both data attribution task and mislablled example identification task.
format Preprint
id arxiv_https___arxiv_org_abs_2508_05089
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Integrated Influence: Data Attribution with Baseline
Yang, Linxiao
Gu, Xinyu
Sun, Liang
Machine Learning
Artificial Intelligence
As an effective approach to quantify how training samples influence test sample, data attribution is crucial for understanding data and model and further enhance the transparency of machine learning models. We find that prevailing data attribution methods based on leave-one-out (LOO) strategy suffer from the local-based explanation, as these LOO-based methods only perturb a single training sample, and overlook the collective influence in the training set. On the other hand, the lack of baseline in many data attribution methods reduces the flexibility of the explanation, e.g., failing to provide counterfactual explanations. In this paper, we propose Integrated Influence, a novel data attribution method that incorporates a baseline approach. Our method defines a baseline dataset, follows a data degeneration process to transition the current dataset to the baseline, and accumulates the influence of each sample throughout this process. We provide a solid theoretical framework for our method, and further demonstrate that popular methods, such as influence functions, can be viewed as special cases of our approach. Experimental results show that Integrated Influence generates more reliable data attributions compared to existing methods in both data attribution task and mislablled example identification task.
title Integrated Influence: Data Attribution with Baseline
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2508.05089