Summary of Eros: Entity-driven Controlled Policy Document Summarization, by Joykirat Singh et al.
EROS: Entity-Driven Controlled Policy Document Summarization
by Joykirat Singh, Sehban Fazili, Rohan Jain, Md Shad Akhtar
First submitted to arxiv on: 29 Feb 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a novel approach to enhancing the readability and interpretability of privacy policy documents by leveraging controlled abstractive summarization techniques. The authors develop PD-Sum, a dataset with labeled privacy-related entities, and design EROS, a model that identifies critical entities through span-based entity extraction and controls summary information content using proximal policy optimization (PPO). The proposed approach demonstrates encouraging improvement over various baselines in terms of entity extraction accuracy and human evaluation. Furthermore, the paper presents qualitative evaluations to establish the effectiveness of EROS. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper helps us understand how we can make privacy policies easier to read. Right now, these policies are often very long and complicated, making it hard for people to know what organizations do with their personal data. The authors created a new way to summarize these policies, which includes important details about what kind of data is being collected and why. This approach uses artificial intelligence to help make the summaries more accurate and helpful. |
Keywords
» Artificial intelligence » Optimization » Summarization
