Explainable Statute Prediction

Overview

Given the factual description of an Indian Supreme Court case, participants must:

1. Identify which sections of the Indian Penal Code (IPC) are applicable
2. Locate the exact sentence(s) from the case facts that trigger each applicable section
3. Explain the legal reasoning connecting each fact sentence to the applicable IPC section

This task tests both legal knowledge (which sections apply?) and interpretability (why do they apply, grounded in the facts?).

Task Description

The task requires predicting applicable IPC sections from case facts, grounding each prediction in specific sentences, and providing legal reasoning that connects the facts to the statute. This is a multi-output structured prediction problem combining classification, extraction, and generation.

Jurisdiction & Scope

Legal categories: Criminal, Civil, Constitutional, Tax, Labor, Commercial, Revenue, Administration, Environmental

Training Data

Release date: 15 June 2026 → 20 June 2026

Each line in the training set has the following structure:

{
  "doc_id": "2013.INSC.228.txt",
  "fact": " The facts which are essential to be stated for adjudication of this appeal are that an FIR was lodged by Prem Singh, PW-2...",
  "explanation": {
    "The facts which are essential to be stated for adjudication of this appeal are that an FIR was lodged by Prem Singh, PW-2, alleging that about 00 p.m.": "IPC 147",
    "on 1992, on hearing a gunshot sound and simultaneously the cry of his brother, Gopal Singh, PW-1, that he was being assaulted and his life was in danger, he rushed to the shop of Gopal Singh and found that accused Gopal Singh and his brother Puran Singh were beating him with hands, fists and stones.": "IPC 147"
  },
  "statute": ["IPC 147"]
}

A single case may have multiple IPC sections in its statute list. The explanation field maps each sentence in the fact to the specific IPC section it supports. Multiple sentences may map to the same statute.

Test Data

Format: JSONL — only doc_id and fact are provided. No statute labels, no explanation sentences.

{
  "doc_id": "2002.INSC.274.txt",
  "fact": "By the impugned judgment the High Court while convicting the respondent of the offence under Section 376 of the Indian Penal Code reduced his sentence..."
}

Participants must predict all three outputs: statute section labels, explanation sentences, and reasoning traces.

Submission Format

Participants submit a single JSONL file. The submission format must contain doc_id and statute (list of predictions):

{
  "doc_id": "2002.INSC.274.txt",
  "statute": [
    {
      "section": "IPC 376",
      "exact_fact": "She recognized the respondent Kishanlal and asked him as to why he had come. He said that he had come to have sexual intercourse with her.",
      "reasoning_trace": "Section 376 IPC applies because the facts establish that the respondent committed sexual intercourse with the prosecutrix against her will, satisfying the ingredients of rape under Section 375 IPC."
    },
    {
      "section": "IPC 457",
      "exact_fact": "While going to Ramlila her husband had bolted the house from outside. At about 11-12 O' clock at night she woke up as someone opened the door.",
      "reasoning_trace": "Section 457 IPC applies because the respondent entered the house as a trespasser with intent to commit an offence, having opened the bolted door to gain entry at night."
    }
  ]
}

Evaluation

Note: Metrics and weights are tentative and may be updated before the test data release.

* Final details — including metric weights, prompt templates, and evaluation pipeline — will be confirmed after 15 June 2026 → 20 June 2026. Minor updates to prompt specifications may occur.

Composite Score = weighted sum of above metrics.

Timeline

Baseline Systems

At least one baseline system will be provided to participants. Potential baselines include:

Ethical Considerations

• All case data is from publicly available Indian Supreme Court judgments
• No personally identifiable information is included
• The task is designed to advance legal AI interpretability, not to provide legal advice