# How well do LLMs reason over tabular data, really?

arXiv: https://arxiv.org/abs/2505.07453 · HTML: https://mlsanity.com/paper/2505.07453/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning.
- Builds on 5 of its 22 references that are in the map.
- First posted on arXiv 12 May 2025.

## Placed in

- LLM reasoning / [Knowledge- and tool-augmented reasoning](https://mlsanity.com/lr/knowledge-and-tool-augmented-reasoning/)

## Builds on (most cited first, 5 of 5)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)

## Built on by (most cited first, 6 of 6)

- [Comprehension Without Competence: Architectural Limits of LLMs in Symbolic Computation and Reasoning](https://mlsanity.com/paper/2507.10624/) (2025)
- [Table Question Answering in the Era of Large Language Models: A Comprehensive Survey of Tasks, Methods, and Evaluation](https://mlsanity.com/paper/2510.09671/) (2025)
- [Merging Continual Pretraining Models for Domain-Specialized LLMs: A Case Study in Finance](https://mlsanity.com/paper/2511.02451/) (2025)
- [Something's Fishy In The Data Lake: A Critical Re-evaluation of Table Union Search Benchmarks](https://mlsanity.com/paper/2505.21329/) (2025)
- [When LLMs Read Tables Carelessly: Measuring and Reducing Data Referencing Errors](https://mlsanity.com/paper/2606.32029/) (2026)
- [LLM4RTL: Tool-Assisted LLM for RTL Generation](https://mlsanity.com/paper/2606.15500/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) excel in natural language tasks, but less is known about their reasoning capabilities over tabular data. Prior analyses devise evaluation strategies that poorly reflect an LLM's realistic performance on tabular queries. Moreover, we have a limited understanding of the robustness of LLMs towards realistic variations in tabular inputs. Therefore, we ask: Can general-purpose LLMs reason over tabular data, really?, and focus on two questions 1) are tabular reasoning capabilities of general-purpose LLMs robust to real-world characteristics of tabular inputs, and 2) how can we realistically evaluate an LLM's performance on analytical tabular queries? Building on a recent tabular reasoning benchmark, we first surface shortcomings of its multiple-choice prompt evaluation strategy, as well as commonly used free-form text metrics such as SacreBleu and BERT-score. We show that an LLM-as-a-judge procedure yields more reliable performance insights and unveil a significant deficit in tabular reasoning performance of LLMs. We then extend the tabular inputs reflecting three common characteristics in practice: 1) missing values, 2) duplicate entities, and 3) structural variations. Experiments show that the tabular reasoning capabilities of general-purpose LLMs suffer from these variations, stressing the importance of improving their robustness for realistic tabular inputs.