Flacuna: Unleashing the Problem Solving Power of Vicuna using FLAN Fine-Tuning

TL;DR

This paper demonstrates that fine-tuning VICUNA with a specialized instruction dataset called FLANMINI significantly enhances its problem-solving capabilities, outperforming other large language models on various benchmarks.

Contribution

The study introduces FLACUNA, a fine-tuned VICUNA model using FLANMINI, showing improved problem-solving performance over existing models.

Findings

FLACUNA outperforms baseline models on multiple benchmarks.

Fine-tuning with FLANMINI enhances problem-solving skills.

VICUNA's performance is significantly improved through targeted instruction tuning.

Abstract

Recently, the release of INSTRUCTEVAL has provided valuable insights into the performance of large language models (LLMs) that utilize encoder-decoder or decoder-only architecture. Interestingly, despite being introduced four years ago, T5-based LLMs, such as FLAN-T5, continue to outperform the latest decoder-based LLMs, such as LLAMA and VICUNA, on tasks that require general problem-solving skills. This performance discrepancy can be attributed to three key factors: (1) Pre-training data, (2) Backbone architecture, and (3) Instruction dataset. In this technical report, our main focus is on investigating the impact of the third factor by leveraging VICUNA, a large language model based on LLAMA, which has undergone fine-tuning on ChatGPT conversations. To achieve this objective, we fine-tuned VICUNA using a customized instruction dataset collection called FLANMINI. This collection includes a subset of the large-scale instruction dataset known as FLAN, as well as various code-related datasets and conversational datasets derived from ChatGPT/GPT-4. This dataset comprises a large number of tasks that demand problem-solving skills. Our experimental findings strongly indicate that the enhanced problem-solving abilities of our model, FLACUNA, are obtained through fine-tuning VICUNA on the FLAN dataset, leading to significant improvements across numerous benchmark datasets in INSTRUCTEVAL. FLACUNA is publicly available at https://huggingface.co/declare-lab/flacuna-13b-v1.0.