Readability score benchmark of the pre- and post-hypernym substituted text using the GPT-J-6B approach for text simplification.

Figshare2024-04-16 更新2026-04-28 收录

下载链接：

https://figshare.com/articles/dataset/Readability_score_benchmark_of_the_pre-_and_post-hypernym_substituted_text_using_the_GPT-J-6B_approach_for_text_simplification_/25612849

下载链接

链接失效反馈

官方服务：

资源简介：

Readability score benchmark of the pre- and post-hypernym substituted text using the GPT-J-6B approach for text simplification.

创建时间：

2024-04-16

Readability score benchmark of the pre- and post-hypernym substituted text using the Programmatic+AI approach for text simplification.

Figshare2024-04-16 更新00

Human-rated benchmark of 3 readability metrics (clarity, conciseness, ease of understanding) between pre- and post-hypernym substituted text generated using the Programmatic+AI, T5, and GPT-J-6B approaches.

Figshare2024-04-16 更新00

Mean change (Δ) in four readability metrics FKS, FKG, ARI, GFI, as well as two sentence complexity metrics MLTD and MDD, of the post-hypernym substituted texts generated using each model compared to the original pre-hypernym substituted texts.

P value of two-sided Wilcoxon Signed Rank test are seen below the mean difference comparing between the post-hypernym substituted texts generated using each model and the original pre-hypernym substit

Figshare2024-04-16 更新00

Readability score benchmark of the pre- and post-hypernym substituted text using the T5 approach for text simplification.

Figshare2024-04-16 更新00

Human-rated benchmark of 5 readability metrics (clarity, conciseness, ease of understanding, accuracy, fidelity) between the post-hypernym substituted text generated using the Programmatic+AI, T5, and GPT-J-6B approaches.

Figshare2024-04-16 更新00