Indian Judiciary is one of the largest Judicial Systems in the world, consisting of the Supreme Court of India, 25 High courts and 72 District courts. All of them produce an enormous amount of legal judgements, some of which run into 100s of pages. Legal practitioners generally depend on manually written summaries, also known as Headnotes, while referring to these judgements. However creating Headnotes takes considerable human effort and is a very slow process. In this AILA task we aim to address this issue by focusing on automatic summarization of legal judgements.

Task Description

Task 2a - Identify "summary-worthy" sentences in a court judgement

Given a court judgement, not all part of the judgement are equally important when creating a headnote. For instance, often the facts and rationale of the judgement are given more importance compared to a precedence while creating a headnote. In task-2a we aim to replicate that behaviour. Given a judgement the task is to identify sentences which are "summary worthy", i.e. they have at least some information which should be included in the summary. This task can be seen as a sentence classification task.

Task 2b - Automatically generate a summary from a given court judgement

Given a court judgement, the participants have to automatically generate a summary, either extractive or abstractive, for it. This subtask can be seen as a continuation of task 2a or as a separate subtask. For instance, the extractive summary could simply be formed by collecting and reordering the sentences identified as important in task 2a. On the other hand these sentences can be compressed/re-written, or generative models can be used to generate abstractive summaries.

The court judgements can vary substantially in length and as a result so will the corresponding headnotes. For each judgement the target summary length in number of words will be provided.