Competition on Legal Information Extraction / Entailment (COLIEE)

A Competition on Legal Information Extraction / Entailment (COLIEE) is an AI competition that focuses on the legal information retrieval, legal information extraction, and legal information entailment.

• Context:
  • It can involve multiple tasks aimed at improving the efficiency and accuracy of legal document retrieval and legal case prediction.
  • It can offer a platform for researchers to apply machine learning and natural language processing techniques to legal texts.
  • It can (typically) include:
    • a Relevant Legal Case Identification Task.
    • a Legal Case Decision Prediction Task.
    • a Legal Article Retrieval Task.
    • a Yes/No Legal Question Answering Task.
  • It can evaluate submissions based on metrics like precision, recall, F-measure (for case law tasks), and F2-measure (for statute law tasks).
  • It can require participants to submit papers detailing their methodologies and experimental results for review and presentation at the associated COLIEE workshop.
  • It can emphasize the importance of reproducibility and the non-use of test datasets for training.
• Example(s):
  • A COLIEE-2020 Competition.
  • A COLIEE-2021 Competition.
  • A COLIEE-2022 Competition.
  • A COLIEE-2023 Competition.
  • A COLIEE-2024 Competition.
  • ...
• Counter-Example(s):
  • A Data Science Competition (not focused on legal texts).
  • A Legal moot court competition.

See: AI Competition, Legal Document Retrieval, Legal Case Prediction.