Summary of Il-tur: Benchmark For Indian Legal Text Understanding and Reasoning, by Abhinav Joshi and Shounak Paul and Akshat Sharma and Pawan Goyal and Saptarshi Ghosh and Ashutosh Modi
IL-TUR: Benchmark for Indian Legal Text Understanding and Reasoning
by Abhinav Joshi, Shounak Paul, Akshat Sharma, Pawan Goyal, Saptarshi Ghosh, Ashutosh Modi
First submitted to arxiv on: 7 Jul 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
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 This paper addresses the pressing need for efficient processing of exponential case growth in legal systems worldwide by proposing IL-TUR, a benchmark for Indian Legal Text Understanding and Reasoning. The benchmark consists of monolingual (English, Hindi) and multilingual tasks addressing different aspects of the legal system from understanding and reasoning perspectives on Indian legal documents. Baseline models, including LLM-based approaches, are presented for each task, highlighting gaps between models and ground truth. A leaderboard is created to foster further research in the legal domain, enabling researchers to upload and compare legal text understanding systems. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps make the legal system more efficient by using special computer programs (NLP and ML) to understand and process a huge number of documents. It’s hard to compare different computer models designed specifically for this task. The paper solves this problem by creating IL-TUR, which has tasks that test how well computers can understand Indian legal documents in English and Hindi, as well as nine other Indian languages. The paper also shows baseline models for each task and what they’re missing compared to the correct answers. This will help researchers make better computer systems for understanding legal texts. |
Keywords
* Artificial intelligence * Nlp
