MRT at IberLEF-2025 PRESTA Task: Maximizing Recovery from Tables with Multiple Steps

TL;DR

This paper introduces a multi-step, LLM-based approach for answering questions about Spanish tables, achieving 85% accuracy by generating and executing Python code to process table data.

Contribution

It presents a novel multi-step method utilizing open-source LLMs and optimized prompts for table question answering in Spanish, building on prior MRT work.

Findings

Achieved 85% accuracy in the PRESTA task

Implemented a multi-step process involving code generation and execution

Utilized open-source LLMs with fine-tuned prompts

Abstract

This paper presents our approach for the IberLEF 2025 Task PRESTA: Preguntas y Respuestas sobre Tablas en Espa\~nol (Questions and Answers about Tables in Spanish). Our solution obtains answers to the questions by implementing Python code generation with LLMs that is used to filter and process the table. This solution evolves from the MRT implementation for the Semeval 2025 related task. The process consists of multiple steps: analyzing and understanding the content of the table, selecting the useful columns, generating instructions in natural language, 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 step. With this approach, we achieved an accuracy score of 85\% in the task.