# Large Language Models for Missing Data Imputation: Understanding Behavior, Hallucination Effects, and Control Mechanisms

arXiv: https://arxiv.org/abs/2603.22332 · HTML: https://mlsanity.com/paper/2603.22332/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 24 references that are in the map.
- First posted on arXiv 20 Mar 2026.

## Builds on (most cited first, 7 of 7)

- [A Survey of Large Language Models](https://mlsanity.com/paper/2303.18223/) (2023)
- [Generative AI](https://mlsanity.com/paper/2309.07930/) (2023)
- [Matrix Completion and Low-Rank SVD via Fast Alternating Least Squares](https://mlsanity.com/paper/1410.2596/) (2014)
- [GATGPT: A Pre-trained Large Language Model with Graph Attention Network for Spatiotemporal Imputation](https://mlsanity.com/paper/2311.14332/) (2023)
- [Data Imputation using Large Language Model to Accelerate Recommendation System](https://mlsanity.com/paper/2407.10078/) (2024)
- [A Context-Aware Approach for Enhancing Data Imputation with Pre-trained Language Models](https://mlsanity.com/paper/2405.17712/) (2024)
- [Table Transformers for Imputing Textual Attributes](https://mlsanity.com/paper/2408.02128/) (2024)

## Built on by (most cited first, 2 of 2)

- [Joint Treatment Effect Estimation from Incomplete Healthcare Data: Temporal Causal Normalizing Flows with LLM-driven Evolutionary MNAR Imputation](https://mlsanity.com/paper/2605.05125/) (2026)
- [Data Language Models: A New Foundation Model Class for Tabular Data](https://mlsanity.com/paper/2605.06290/) (2026)

## Abstract (arXiv)

Data imputation is a cornerstone technique for handling missing values in real-world datasets, which are often plagued by missingness. Despite recent progress, prior studies on Large Language Models-based imputation remain limited by scalability challenges, restricted cross-model comparisons, and evaluations conducted on small or domain-specific datasets. Furthermore, heterogeneous experimental protocols and inconsistent treatment of missingness mechanisms (MCAR, MAR, and MNAR) hinder systematic benchmarking across methods. This work investigates the robustness of Large Language Models for missing data imputation in tabular datasets using a zero-shot prompt engineering approach. To this end, we present a comprehensive benchmarking study comparing five widely used LLMs against six state-of-the-art imputation baselines. The experimental design evaluates these methods across 29 datasets (including nine synthetic datasets) under MCAR, MAR, and MNAR mechanisms, with missing rates of up to 20%. The results demonstrate that leading LLMs, particularly Gemini 3.0 Flash and Claude 4.5 Sonnet, consistently achieve superior performance on real-world open-source datasets compared to traditional methods. However, this advantage appears to be closely tied to the models' prior exposure to domain-specific patterns learned during pre-training on internet-scale corpora. In contrast, on synthetic datasets, traditional methods such as MICE outperform LLMs, suggesting that LLM effectiveness is driven by semantic context rather than purely statistical reconstruction. Furthermore, we identify a clear trade-off: while LLMs excel in imputation quality, they incur significantly higher computational time and monetary costs. Overall, this study provides a large-scale comparative analysis, positioning LLMs as promising semantics-driven imputers for complex tabular data.