arXiv：cs.LG· Anuraj Maurya·· 4 小时前AI 评分34
Mamba 与 Transformer 法律 AI 基准测试：法条分类与判例检索
Scaling Legal AI: Benchmarking Mamba and Transformers for Statutory Classification and Case Law Retrieval
AI 导读
一项预印本基准测试对比了 Mamba、SSD-Mamba 与 BERT、DeBERTa、Longformer 在法律分类和判例检索任务上的表现。
正文
Abstract:Statutory corpora and judicial decisions are growing faster than legal professionals can read them, while individual judgments often exceed the context limits of standard encoder models. Transformer architectures dominate legal NLP benchmarks, but their quadratic attention complexity can require truncating or fragmenting documents that demand whole-document reasoning. Selective state-space models (SSMs), such as Mamba, offer linear-time sequence modeling and are a promising alternative for long legal documents, yet their performance on legal classification and retrieval remains underexplored.
We present a preliminary benchmark comparing Mamba and SSD-Mamba with BERT, DeBERTa, and Longformer across four legal classification tasks (ECtHR, EUR-Lex, SCOTUS, and ILDC/ILC) and two case-retrieval tasks (ECtHR and ILDC), using a shared windowing and aggregation pipeline. The strongest SSM performs within approximately 1.3 percentage points of the strongest transformer across tasks and metrics. SSD-Mamba achieves the best results on most metrics for ECtHR classification, ILDC classification, and ECtHR retrieval, while processing approximately 3 times more tokens per second than DeBERTa and 4 times more than Longformer. DeBERTa remains strongest on SCOTUS and EUR-Lex F1.
These results are preliminary because they do not include variance estimates across random seeds or statistical significance testing. Rather than presenting a definitive ranking, we use these findings to motivate further evaluation with repeated-seed experiments, statistical testing, and controls for model capacity and computational efficiency.
| Subjects: | Computers and Society (cs.CY); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2509.00141 [cs.CY] |
| (or arXiv:2509.00141v2 [cs.CY] for this version) | |
| https://doi.org/10.48550/arXiv.2509.00141 arXiv-issued DOI via DataCite |
Submission history
From: Anuraj Maurya [view email]
[v1]
Fri, 29 Aug 2025 17:38:47 UTC (81 KB)
[v2]
Wed, 7 Oct 2026 15:52:00 UTC (143 KB)
来源：arXiv：cs.LG · arxiv.org