Sheffield NLP at FinCausal 2026: A comparative study of RAG approaches and fine-tuning for causal Q&A in financial texts

This paper describes our approach to the FinCausal 2026 shared task, which addresses causal question answering from financial documents in English and Spanish. We investigated the effectiveness of fine-tuned generative models combined with Retrieval-Augmented Generation (RAG). Our approach compares five retrieval strategies across base and fine-tuned GPT models (GPT-4.1-mini). RAG-based few-shot selection performed better than random sampling, particularly for the base model. In the FinCausal 2026 official run, this approach was ranked first in both the English and Spanish sub-tasks, obtaining LLM scores of 4.8140 and 4.8131 out of 5, respectively.