研究法律文本随时间演变对判决预测的影响，发现模型会严重退化。
Temporal Concept Drift in Legal Judgment Prediction: Neural Baselines Across Three Epochs of Ukrainian Court Decisions
- 用三个历史时期数据训练和测试模型，构建跨时序评估矩阵。
- 战前数据训练的模型在战时预测中准确率下降27.2个百分点。
- 法律领域预训练可缓解退化，但无法消除时间漂移的本质影响。
法律NLP基准通常在随机划分的数据上评估模型，隐含假设法律语言是静态的。本文通过在乌克兰法院判决数据的三个时间阶段——战前（2008–2013）、混合战争（2014–2021）和全面入侵（2022–2026）——上微调四种Transformer编码器（XLM-RoBERTa base/large及其法律领域变体），检验该假设。每个模型在一个时期训练，在所有三个时期上评估，形成3×3跨时序泛化矩阵。四项发现：（1）正向退化严重：战前训练模型在全面入侵期判决预测中宏平均F1下降高达27.2个百分点；（2）退化不对称：从战时回溯到战前表现更稳定，支持法律语言具有累积性的假说；（3）法律领域预训练（Legal-XLM-R）未提升绝对性能，但减轻了正向退化幅度与不对称性；（4）按时间顺序持续学习可完全保留战前知识（+1.8至+6.2个百分点），同时提升战时性能（+16.5至+19.0个百分点）；逆序训练导致严重遗忘。跨司法管辖区在瑞士判决预测数据上预训练可提升绝对性能，但不减少时间退化，确认时间漂移是法律语言演化的内在属性。数据集（共42.8万份判决）作为LEXTREME贡献公开发布。
原文摘要 · Abstract (English)
Legal NLP benchmarks evaluate models on randomly split data, implicitly assuming that legal language is stationary. We test this assumption by fine-tuning four transformer encoders -- XLM-RoBERTa (base and large) and their legal-domain variants -- on Ukrainian court decisions from three temporal epochs defined by geopolitical disruptions: pre-war (2008-2013), hybrid war (2014-2021), and full-scale invasion (2022-2026). Each model is trained on one epoch and evaluated on all three, producing a 3x3 cross-temporal generalization matrix. Four findings emerge. (1) Forward degradation is severe: models trained on pre-war data lose up to 27.2 percentage points of macro-F1 when applied to full-scale invasion era decisions. (2) The degradation is asymmetric: backward transfer (full-scale to pre-war) is substantially more robust than forward transfer, consistent with the hypothesis that legal language is additive. (3) Legal-domain pretraining (Legal-XLM-R) does not improve absolute performance but reduces forward degradation magnitude and asymmetry. (4) Chronological continual learning eliminates catastrophic forgetting for general XLM-R: pre-war knowledge is fully retained (+1.8 to +6.2 pp) while full-scale performance gains +16.5 to +19.0 pp; reverse-chronological training causes severe forgetting. Cross-jurisdictional pretraining on Swiss Judgment Prediction data improves absolute performance but does not reduce temporal degradation magnitude, confirming that temporal drift is an intrinsic property of legal language evolution. The dataset (428K decisions across three epochs) is publicly available as a LEXTREME contribution.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。