QR-LJP incorporates legal quantitative reasoning into judgment prediction. It extracts monetary values from case facts, determines the total crime amount using legal reasoning, and uses this measure of crime severity to improve judgment predictions, particularly prison terms.

Publication

Published in ACM Transactions on Knowledge Discovery from Data (TKDD) on May 19, 2026. Volume 20, Issue 4, Article 66, pages 1–42.

DOI: 10.1145/3807945.

Overview

Monetary features can be critical for assessing crime severity, yet models may overlook their legal significance. QR-LJP uses a domain-adapted language model to extract relevant amounts and perform quantitative reasoning. The resulting crime amount is aligned with legal knowledge to support judgment prediction.

Evaluation

The study evaluates legal judgment prediction on CAIL-2018 and examines how quantitative reasoning improves existing methods, with particular attention to prison term prediction.

Resources