Summary of A Comparative Analysis Of Instruction Fine-tuning Llms For Financial Text Classification, by Sorouralsadat Fatemi et al.
A Comparative Analysis of Instruction Fine-Tuning LLMs for Financial Text Classification
by Sorouralsadat Fatemi, Yuheng Hu, Maryam Mousavi
First submitted to arxiv on: 4 Nov 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This study explores the potential of instruction fine-tuning smaller-scale Large Language Models (LLMs) like Mistral-7B, Llama3-8B, and Phi3-mini to improve their performance in financial text classification tasks. The researchers fine-tuned both instruction-tuned and base models across four financial classification tasks, achieving significant improvements in task-specific performance. They also evaluated the zero-shot capabilities of these fine-tuned models on three unseen complex financial tasks, including argument classification, deal completeness classification, and causal classification. Their results show that while base model fine-tuning led to greater degradation, instruction-tuned models maintained more robust performance. To address this degradation, they employed model merging techniques, integrating single-task domain-specific fine-tuned models with the base model. This merging method resulted in significant enhancements in zero-shot performance, even exceeding the original model’s accuracy on certain datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study looks at how to make language models better at understanding financial texts. They tried making smaller language models work better by giving them instructions specific to financial tasks. They found that this helped a lot and made the models much more accurate. But, they also saw that when they used these new models for other tasks, their performance got worse. To fix this, they combined different versions of the model together. This combination worked even better than the original model on some tasks! The study shows that giving language models specific instructions can help them do better in financial tasks. |
Keywords
» Artificial intelligence » Classification » Fine tuning » Text classification » Zero shot
