# MRT at SemEval-2025 Task 8: Maximizing Recovery from Tables with Multiple Steps

arXiv: https://arxiv.org/abs/2505.22264 · HTML: https://mlsanity.com/paper/2505.22264/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 15 references that are in the map.
- First posted on arXiv 28 May 2025.

## Builds on (most cited first, 10 of 10)

- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [Phi-4 Technical Report](https://mlsanity.com/paper/2412.08905/) (2024)
- [TAPAS: Weakly Supervised Table Parsing via Pre-training](https://mlsanity.com/paper/2004.02349/) (2020)
- [TabFact: A Large-scale Dataset for Table-based Fact Verification](https://mlsanity.com/paper/1909.02164/) (2019)
- [Chain-of-Table: Evolving Tables in the Reasoning Chain for Table Understanding](https://mlsanity.com/paper/2401.04398/) (2024)
- [A Comprehensive Exploration on WikiSQL with Table-Aware Word Contextualization](https://mlsanity.com/paper/1902.01069/) (2019)
- [TableRAG: Million-Token Table Understanding with Language Models](https://mlsanity.com/paper/2410.04739/) (2024)
- [Evaluating and Aligning CodeLLMs on Human Preference](https://mlsanity.com/paper/2412.05210/) (2024)
- [Language Modeling on Tabular Data: A Survey of Foundations, Techniques and Evolution](https://mlsanity.com/paper/2408.10548/) (2024)
- [Accurate and Regret-aware Numerical Problem Solver for Tabular Question Answering](https://mlsanity.com/paper/2410.12846/) (2024)

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

- [ExpliCIT-QA: Explainable Code-Based Image Table Question Answering](https://mlsanity.com/paper/2507.11694/) (2025)
- [MRT at IberLEF-2025 PRESTA Task: Maximizing Recovery from Tables with Multiple Steps](https://mlsanity.com/paper/2507.12981/) (2025)

## Abstract (arXiv)

In this paper we expose our approach to solve the textitSemEval 2025 Task 8: Question-Answering over Tabular Data challenge. Our strategy leverages Python code generation with LLMs to interact with the table and get the answer to the questions. The process is composed of multiple steps: understanding the content of the table, generating natural language instructions in the form of steps to follow in order to get the answer, translating these instructions to code, running it and handling potential errors or exceptions. These steps use open source LLMs and fine grained optimized prompts for each task (step). With this approach, we achieved a score of 70.50% for subtask 1.