LeakageDetector 2.0: Analyzing Data Leakage in Jupyter-Driven Machine Learning Pipelines

Owen Truong 1 , Terrence Zhang 1 , Arnav Marchareddy 1 , Ryan Lee 1 , Jeffery Busold 1 , Michael Socas 1 , Eman Abdullah AlOmar 1 · 2025

Read LeakageDetector 2.0: Analyzing Data Leakage in Jupyter-Driven Machine Learning Pipelines by Owen Truong 1 , Terrence Zhang 1 , Arnav Marchareddy 1 , Ryan Lee 1 , Jeffery Busold 1 , Michael Socas 1 , Eman Abdullah AlOmar 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-15971.txt.

Description

Owen Truong leads a team that builds a tool to catch data leakage in machine learning notebooks. Leakage means test data sneak into training, which inflates scores and misleads builders. It takes three main forms: overlap, preprocessing leaks, and repeated test use. Their tool is an editor extension for the VS Code platform.

The extension scans Jupyter notebooks for these leak types as the user writes. Beyond flagging, it offers two fixes. A quick fix applies a standard manual repair. A language model guide instead walks the developer toward best practice for the pipeline. The aim is cleaner code and honest scores. The paper presents version two of the tool.

Catalog record

CREATOROwen Truong 1 , Terrence Zhang 1 , Arnav Marchareddy 1 , Ryan Lee 1 , Jeffery Busold 1 , Michael Socas 1 , Eman Abdullah AlOmar 1
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE24 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-15971

Browse the full library · About the Tower of Babel