Evaluating Text Summaries Generated by Large Language Models Using OpenAI's GPT
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910436752883712 |
|---|---|
| author | Shakil, Hassan Mahi, Atqiya Munawara Nguyen, Phuoc Ortiz, Zeydy Mardini, Mamoun T. |
| author_facet | Shakil, Hassan Mahi, Atqiya Munawara Nguyen, Phuoc Ortiz, Zeydy Mardini, Mamoun T. |
| contents | This research examines the effectiveness of OpenAI's GPT models as independent evaluators of text summaries generated by six transformer-based models from Hugging Face: DistilBART, BERT, ProphetNet, T5, BART, and PEGASUS. We evaluated these summaries based on essential properties of high-quality summary - conciseness, relevance, coherence, and readability - using traditional metrics such as ROUGE and Latent Semantic Analysis (LSA). Uniquely, we also employed GPT not as a summarizer but as an evaluator, allowing it to independently assess summary quality without predefined metrics. Our analysis revealed significant correlations between GPT evaluations and traditional metrics, particularly in assessing relevance and coherence. The results demonstrate GPT's potential as a robust tool for evaluating text summaries, offering insights that complement established metrics and providing a basis for comparative analysis of transformer-based models in natural language processing tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_04053 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Evaluating Text Summaries Generated by Large Language Models Using OpenAI's GPT Shakil, Hassan Mahi, Atqiya Munawara Nguyen, Phuoc Ortiz, Zeydy Mardini, Mamoun T. Computation and Language Artificial Intelligence Machine Learning This research examines the effectiveness of OpenAI's GPT models as independent evaluators of text summaries generated by six transformer-based models from Hugging Face: DistilBART, BERT, ProphetNet, T5, BART, and PEGASUS. We evaluated these summaries based on essential properties of high-quality summary - conciseness, relevance, coherence, and readability - using traditional metrics such as ROUGE and Latent Semantic Analysis (LSA). Uniquely, we also employed GPT not as a summarizer but as an evaluator, allowing it to independently assess summary quality without predefined metrics. Our analysis revealed significant correlations between GPT evaluations and traditional metrics, particularly in assessing relevance and coherence. The results demonstrate GPT's potential as a robust tool for evaluating text summaries, offering insights that complement established metrics and providing a basis for comparative analysis of transformer-based models in natural language processing tasks. |
| title | Evaluating Text Summaries Generated by Large Language Models Using OpenAI's GPT |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2405.04053 |