MRT at SemEval-2025 Task 8: Maximizing Recovery from Tables with Multiple Steps
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912400053108736 |
|---|---|
| author | Lagos, Maximiliano Hormazábal Saez, Álvaro Bueno Cerezo-Costas, Héctor Doval, Pedro Alonso Vesteiro, Jorge Alcalde |
| author_facet | Lagos, Maximiliano Hormazábal Saez, Álvaro Bueno Cerezo-Costas, Héctor Doval, Pedro Alonso Vesteiro, Jorge Alcalde |
| contents | In this paper we expose our approach to solve the \textit{SemEval 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_22264 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MRT at SemEval-2025 Task 8: Maximizing Recovery from Tables with Multiple Steps Lagos, Maximiliano Hormazábal Saez, Álvaro Bueno Cerezo-Costas, Héctor Doval, Pedro Alonso Vesteiro, Jorge Alcalde Computation and Language Artificial Intelligence Information Retrieval In this paper we expose our approach to solve the \textit{SemEval 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. |
| title | MRT at SemEval-2025 Task 8: Maximizing Recovery from Tables with Multiple Steps |
| topic | Computation and Language Artificial Intelligence Information Retrieval |
| url | https://arxiv.org/abs/2505.22264 |