In this competition, participants are tasked with the Supervised-Fine Tuning(SFT) of Llama3-8B model to predict the correctness of answers to math questions. The goal is to assess whether the provided answer to each question is correct or not.

Supervised fine-tuning of Meta's LLaMA 3.1 8B-Instruct on the RAG-Instruct dataset to enhance retrieval-augmented generation (RAG) capabilities for knowledge-intensive QA.