The author proposes EROS, a model for controlled policy document summarization, to enhance interpretability and readability by including critical privacy-related entities. The approach integrates entity extraction and reinforcement learning to optimize the relevance of generated summaries.
Enhancing interpretability and readability of policy documents through controlled abstractive summarization, focusing on critical privacy-related entities and organization's rationale.
Enhancing privacy policy document interpretability and readability through controlled abstractive summarization.