Learning Fine-Grained Fact-Article Correspondence in Legal Cases

  title={Learning Fine-Grained Fact-Article Correspondence in Legal Cases},
  author={Jidong Ge and Yunyun Huang and Xiaoyu Shen and Chuanyi Li and Wei Hu and Bin Luo},
  journal={IEEE/ACM Transactions on Audio, Speech, and Language Processing},
  • Jidong Ge, Yunyun Huang, B. Luo
  • Published 21 April 2021
  • Computer Science
  • IEEE/ACM Transactions on Audio, Speech, and Language Processing
Automatically recommending relevant law articles to a given legal case has attracted much attention as it can greatly release human labor from searching over the large database of laws. However, current researches only support coarse-grained recommendation where all relevant articles are predicted as a whole without explaining which specific fact each article is relevant with. Since one case can be formed of many supporting facts, traversing over them to verify the correctness of recommendation… 

Dependency Learning for Legal Judgment Prediction with a Unified Text-to-Text Transformer

This work proposes leveraging a unified text-to-text Transformer for LJP, where the dependencies among sub-tasks can be naturally established within the auto-regressive decoder, and shows that this unified transformer, albeit pretrained on general-domain text, outperforms pretrained models tailored specifically for the legal domain.

Unsupervised Simplification of Legal Texts

An unsupervised simplification method for legal texts (USLT), which demonstrates that USLT outperforms state-of-the-art domain-general TS methods in text simplicity while keeping the semantics intact.

A Survey on Legal Judgment Prediction: Datasets, Metrics, Models and Challenges

Up-to-date andhensive review of existing LJP tasks, data sets, models andevaluations are provided to help researchers and legal professionals understand the status of LJP.

Legal Judgment Prediction: A Survey of the State of the Art

Automatic legal judgment prediction (LJP) has recently received increasing attention in the natural language processing community in part because of its practical values as well as the associated



Legal Judgment Prediction via Topological Learning

This work formalizes the dependencies among subtasks as a Directed Acyclic Graph (DAG) and proposes a topological multi-task learning framework, TopJudge, which incorporates multiple subtasks and DAG dependencies into judgment prediction.

Distinguish Confusing Law Articles for Legal Judgment Prediction

An end-to-end model, LADAN, is presented and a novel graph neural network, GDL, is proposed to automatically learn subtle differences between confusing law articles, and also a novel attention mechanism that fully exploits the learned differences to attentively extract effective discriminative features from fact descriptions.

Charge Prediction with Legal Attention

An attentional neural network, LegalAtt, which uses relevant articles to improve the performance and interpretability of charge prediction task, and achieves the best performance on the real-world dataset compared with other state-of-the-art baselines.

Legal Judgment Prediction via Multi-Perspective Bi-Feedback Network

This paper designs a multi-perspective forward prediction and backward verification framework to utilize result dependencies among multiple subtasks effectively and integrates word collocations features of fact descriptions into the network via an attention mechanism.

Neural Legal Judgment Prediction in English

A broad variety of neural models are evaluated on the new English legal judgment prediction dataset, establishing strong baselines that surpass previous feature-based models in three tasks: binary violation classification; multi-label classification; and case importance prediction.

Modeling Dynamic Pairwise Attention for Crime Classification over Legal Articles

This paper treats the evidences as documents, and articles as labels, thus the conviction process can be cast as a multi-label classification problem, and proposes a unified multi-task learning paradigm, DPAM, which outperforms state-of-the-art multi- label classification methods.

Predicting judicial decisions of the European Court of Human Rights: a Natural Language Processing perspective

The empirical analysis indicates that the formal facts of a case are the most important predictive factor, consistent with the theory of legal realism suggesting that judicial decision-making is significantly affected by the stimulus of the facts.

Learning to Predict Charges for Criminal Cases with Legal Basis

It is argued that relevant law articles play an important role in this task, and therefore an attention-based neural network method is proposed to jointly model the charge prediction task and the relevant article extraction task in a unified framework.

Hierarchical Matching Network for Crime Classification

Experiments on two real-world judicial datasets demonstrate that the proposed Hierarchical Matching Network model can significantly outperform the state-of-the-art methods.

Improving Latent Alignment in Text Summarization by Generalizing the Pointer Generator

This paper allows the model to “edit” pointed tokens instead of always hard copying them, and shows the model is able to capture more latent alignment relations than exact word matches, improve word alignment accuracy, and generate higher-quality summaries validated by both qualitative and quantitative evaluations.