Summary of Applicability Of Large Language Models and Generative Models For Legal Case Judgement Summarization, by Aniket Deroy et al.
Applicability of Large Language Models and Generative Models for Legal Case Judgement Summarization
by Aniket Deroy, Kripabandhu Ghosh, Saptarshi Ghosh
First submitted to arxiv on: 6 Jul 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 A new study explores the application of generative models, including abstractive summarization models and Large Language Models (LLMs), for automatic summarization of legal case judgements. The researchers compared these models with extractive summarization methods on two datasets of UK and Indian Supreme Court cases, as well as a third dataset of US Government reports. The results show that generative models generally perform better than extractive methods in terms of traditional summary quality metrics. However, the study also highlights inconsistencies and hallucinations in the outputs of these models, suggesting the need for further improvements to enhance their reliability. Currently, a human-in-the-loop technique is more suitable for identifying inconsistencies in generated summaries. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A team of researchers looked at ways to use computer programs to summarize long legal documents. They tested different types of models that can generate text and compared them with ones that just pick out important parts. The models were tried on three sets of documents: Supreme Court cases from the UK and India, and Government reports from the US. The results showed that the generating models did a better job than the picking-out-parts models at summarizing the documents. But the study also found some problems with the generated summaries, like parts that weren’t really there to begin with. This means that for now, it’s still best to have a person check over the computer-generated summaries. |
Keywords
» Artificial intelligence » Summarization
