CACTI: Leveraging Copy Masking and Contextual Information to Improve Tabular Data Imputation

2025/06/02 by Aditya Gorla, Gorla, Aditya, Qi Wang +7 · 1 citation
Computer Science · #Data Visualization and Analytics #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Time Series Analysis and Forecasting

paper · doi:10.48550/arxiv.2506.02306

openalex publication_date 2025/06/02 · openalex created_date 2025/10/14 · openalex updated_date 2026/08/01

Abstract

gain of 7.8% over the next best method (13.4%, 6.1%, and 5.3% under missing not at random, at random and completely at random, respectively) - across a diverse range of datasets and missingness conditions. Our results highlight the value of leveraging dataset-specific contextual information and missingness patterns to enhance imputation performance. Code is publicly available at github.com/sriramlab/CACTI.

Citations

• mGTE: Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrieval
• Large Scale Transfer Learning for Tabular Data via Language Modeling
• CTSyn: A Foundation Model for Cross Tabular Data Generation
• ReMasker: Imputing Tabular Data with Masked Autoencoding
• UniTabE: A Universal Pretraining Protocol for Tabular Foundation Model in Data Science
• A Complete Characterisation of Structured Missingness
• Do We Still Need Clinical Language Models?
• Learning from data with structured missingness
• Diffusion models for missing value imputation in tabular data
• Multiple Imputation via Generative Adversarial Network for High-dimensional Blockwise Missing Value Problems
• DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing
• Masked Autoencoders Are Scalable Vision Learners
• Position Information in Transformers: An Overview
• Position Information in Transformers: An Overview
• TabTransformer: Tabular Data Modeling Using Contextual Embeddings
• not-MIWAE: Deep Generative Modelling with Missing not at Random Data
• TaBERT: Pretraining for Joint Understanding of Textual and Tabular Data
• Missing Data Imputation using Optimal Transport
• TabNet: Attentive Interpretable Tabular Learning
• Publicly Available Clinical BERT Embeddings
• MIWAE: Deep Generative Modelling and Imputation of Incomplete Data
• Handling Incomplete Heterogeneous Data using VAEs
• GAIN: Missing Data Imputation using Generative Adversarial Nets
• Matrix Completion and Low-Rank SVD via Fast Alternating Least Squares
• mice: Multivariate Imputation by Chained Equations inR
• DiffPuter: Empowering Diffusion Models for Missing Data Imputation

Cited by

• TabImpute: Universal Zero-Shot Imputation for Tabular Data

Related

• ReMasker: Imputing Tabular Data with Masked Autoencoding
• To Predict or Not To Predict? Proportionally Masked Autoencoders for Tabular Data Imputation
• Not Another Imputation Method: A Transformer-based Model for Missing Values in Tabular Datasets
• Unmasking Trees for Tabular Data
• TabINR: An Implicit Neural Representation Framework for Tabular Data Imputation