Generative AI-Based Text Generation Methods Using Pre-Trained GPT-2 Model

TL;DR

This paper compares various text generation techniques using the pre-trained GPT-2 model, analyzing their strengths, weaknesses, and applications through a comprehensive evaluation and suggesting future research directions.

Contribution

It provides a comparative analysis of multiple text generation methods with insights into their performance and potential improvements.

Findings

Different methods have distinct strengths and weaknesses.

Performance varies significantly across techniques.

The study highlights promising future research directions.

Abstract

This work delved into the realm of automatic text generation, exploring a variety of techniques ranging from traditional deterministic approaches to more modern stochastic methods. Through analysis of greedy search, beam search, top-k sampling, top-p sampling, contrastive searching, and locally typical searching, this work has provided valuable insights into the strengths, weaknesses, and potential applications of each method. Each text-generating method is evaluated using several standard metrics and a comparative study has been made on the performance of the approaches. Finally, some future directions of research in the field of automatic text generation are also identified.