Summary of From Graph to Word Bag: Introducing Domain Knowledge to Confusing Charge Prediction, by Ang Li et al.
From Graph to Word Bag: Introducing Domain Knowledge to Confusing Charge Prediction
by Ang Li, Qiangchao Chen, Yiquan Wu, Ming Cai, Xiang Zhou, Fei Wu, Kun Kuang
First submitted to arxiv on: 7 Mar 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Computation and Language (cs.CL)
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 A novel approach to predicting confusing charges in legal AI is introduced, which leverages domain knowledge and attention mechanisms to improve model performance. The From Graph to Word Bag (FWGB) method constructs a legal knowledge graph containing constituent elements that guide the selection of keywords for each charge. This word bag is used to expand the attention mechanism, introducing a new loss function with attention supervision through words in the word bag. The approach demonstrates exceptional performance, particularly in imbalanced label distributions, and outperforms existing methods on a dataset constructed from real-world judicial documents. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In this paper, researchers are trying to make computers better at understanding what’s happening when people commit crimes. They’re working on a special problem called “confusing charges,” where it’s hard to decide which charge is the right one because of subtle differences between them. The new approach they came up with uses something like a dictionary or a set of rules to help the computer figure out what’s important in each case. It seems to work really well, especially when there are more cases on one side than another. |
Keywords
» Artificial intelligence » Attention » Knowledge graph » Loss function
