Utilizing GPT to Enhance Text Summarization: A Strategy to Minimize Hallucinations

arXiv:2405.04039v1 Announce Type: cross
Abstract: In this research, we uses the DistilBERT model to generate extractive summary and the T5 model to generate abstractive summaries. Also, we generate hybrid summaries by combining both DistilBERT and T5 models. Central to our research is the implementation of GPT-based refining process to minimize the common problem of hallucinations that happens in AI-generated summaries. We evaluate unrefined summaries and, after refining, we also assess refined summaries using a range of traditional and novel metrics, …

abstract arxiv cs.ai cs.cl cs.lg distilbert generate gpt hallucinations hybrid implementation process research strategy summarization summary text text summarization type