LLI-UAM Team at FinancES 2023: Noise, Data Augmentation and Hallucinations

Indexed in

License and use

Citations

Altmetrics

Analysis of institutional authors

Porta-Zamorano JAuthorMoreno-Sandoval AAuthor

November 6, 2023

Publications

Proceedings Paper

LLI-UAM Team at FinancES 2023: Noise, Data Augmentation and Hallucinations

Publicated to:CEUR Workshop Proceedings. 3496 - 2023-01-01 3496(), DOI:

Authors: Porta-Zamorano J; Torterolo Y; Moreno-Sandoval A

Affiliations

Universidad Autónoma de Madrid - Author

Abstract

This paper describes the T5-based system developed for FinancES 2023 Shared Task by the Laboratorio de Lingüística Informática at UAM. The LLI-UAM system achieved a good ranking in all the tasks. The paper also describes some noise and data augmentation or hallucination mitigation experiments. In particular, we used corrected versions of the datasets to evaluate the impact of noise. Moreover, ChatGPT was utilised to augment the data and improve accuracy in tagging. We also describe the presence of hallucinations. Ultimately, we identify the best model for each task and draw conclusions based on our findings.

Keywords

ChatgptData augmentationFinances shared taskHallucinationsMt5Noise

Quality index

Bibliometric impact. Analysis of the contribution and dissemination channel

The work has been published in the journal CEUR Workshop Proceedings, Q4 Agency Scopus (SJR), its regional focus and specialization in Computer Science (Miscellaneous), give it significant recognition in a specific niche of scientific knowledge at an international level.

Leadership analysis of institutional authors

There is a significant leadership presence as some of the institution’s authors appear as the first or last signer, detailed as follows: First Author (PORTA ZAMORANO, JORDI) and Last Author (MORENO SANDOVAL, ANTONIO).