Summary of Automated Literature Review Using Nlp Techniques and Llm-based Retrieval-augmented Generation, by Nurshat Fateh Ali et al.
Automated Literature Review Using NLP Techniques and LLM-Based Retrieval-Augmented Generation
by Nurshat Fateh Ali, Md. Mahdi Mohtasim, Shakil Mosharrof, T. Gopi Krishna
First submitted to arxiv on: 27 Nov 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR); Machine Learning (cs.LG)
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 paper explores multiple approaches to automate literature review generation using NLP techniques and retrieval-augmented generation with Large Language Models (LLMs). The goal is to develop a system that can generate reviews from PDF files as input, addressing the challenge of manually reviewing ever-increasing research articles. Various NLP strategies are evaluated, including frequency-based methods like spaCy, transformer models like Simple T5, and retrieval-augmented generation with GPT-3.5-turbo. The SciTLDR dataset is used to implement three distinct systems for auto-generating reviews, with ROUGE scores serving as the evaluation metric. The results show that Large Language Model GPT-3.5-turbo achieves the highest ROUGE-1 score of 0.364, followed by the transformer model and spaCy. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper tries to make computers do some work that humans usually do – reading and summarizing lots of research articles. Right now, people have to read many papers to understand what’s already known on a topic. This can be very time-consuming. The researchers want to create a computer program that can read these papers and summarize them automatically. They tested different ways for the computer to do this using special language processing techniques and a really powerful language model called GPT-3.5-turbo. They found that this approach worked best, but it’s not perfect yet. |
Keywords
» Artificial intelligence » Gpt » Language model » Large language model » Nlp » Retrieval augmented generation » Rouge » T5 » Transformer
