Trustworthiness of Children Stories Generated by Large Language Models
Bhandari, Prabin, Brennan, Hannah Marie
–arXiv.org Artificial Intelligence
LLMs' ability to generate age-appropriate Advancements in pretrained large language models materials to target audiences also becomes a vital (LLMs) like GPT-3 (Brown et al., 2020) and aspect of overall trustworthiness. LLaMA (Touvron et al., 2023), have made it easier To assess the trustworthiness of LLMs in generating to generate natural language text for a variety of children's stories, we use two open-source downstream tasks, including generating narrative foundation language models, OPT (Zhang et al., text like children's stories. The ability to generate 2022) and LLaMA (Touvron et al., 2023), along natural text using LLMs has seen substantial with an instruction-following model Alpaca (Taori improvement with the innovation of instructionfollowing et al., 2023) to generate children's stories. Then, models like InstructGPT (Ouyang et al., we compare these generated stories against actual 2022) and Alpaca (Taori et al., 2023), resulting in children's stories, old and modern.
arXiv.org Artificial Intelligence
Jul-25-2023
- Country:
- North America > United States
- New York > New York County
- New York City (0.04)
- Minnesota > Hennepin County
- Minneapolis (0.14)
- New York > New York County
- Europe > Finland
- Asia > Japan
- Honshū > Kantō > Tokyo Metropolis Prefecture > Tokyo (0.14)
- North America > United States
- Genre:
- Research Report > New Finding (1.00)
- Technology: