ChatGPT vs Human-authored Text: Insights into Controllable Text Summarization and Sentence Style Transfer
This work addresses the problem of assessing the limitations of large language models in generating stylistically varied and faithful text for researchers and practitioners in NLP, but it is incremental as it builds on existing evaluations of such models.
The paper systematically evaluated ChatGPT's performance in controllable text generation tasks, specifically for adapting output to different audiences and writing styles, and found that human-authored texts exhibit greater stylistic variation and that ChatGPT sometimes introduces factual errors or hallucinations.
Large-scale language models, like ChatGPT, have garnered significant media attention and stunned the public with their remarkable capacity for generating coherent text from short natural language prompts. In this paper, we aim to conduct a systematic inspection of ChatGPT's performance in two controllable generation tasks, with respect to ChatGPT's ability to adapt its output to different target audiences (expert vs. layman) and writing styles (formal vs. informal). Additionally, we evaluate the faithfulness of the generated text, and compare the model's performance with human-authored texts. Our findings indicate that the stylistic variations produced by humans are considerably larger than those demonstrated by ChatGPT, and the generated texts diverge from human samples in several characteristics, such as the distribution of word types. Moreover, we observe that ChatGPT sometimes incorporates factual errors or hallucinations when adapting the text to suit a specific style.