English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Pre-trained Transformers currently dominate most NLP tasks. They impose,however, limits on the maximum input length (512 sub-words in BERT), which aretoo restrictive in the legal domain. Even sparse-attention models, such asLongformer and BigBird, which increase the maximum input length to 4,096sub-words, severely truncate texts in three of the six datasets of LexGLUE.Simpler linear classifiers with TF-IDF features can handle texts of any length,require far less resources to train and deploy, but are usually outperformed bypre-trained Transformers. We explore two directions to cope with long legaltexts: (i) modifying a Longformer warm-started from LegalBERT to handle evenlonger texts (up to 8,192 sub-words), and (ii) modifying LegalBERT to useTF-IDF representations. The first approach is the best in terms of performance,surpassing a hierarchical version of LegalBERT, which was the previous state ofthe art in LexGLUE. The second approach leads to computationally more efficientmodels at the expense of lower performance, but the resulting models stilloutperform overall a linear SVM with TF-IDF features in long legal documentclassification.

Processing Long Legal Documents with Pre-trained Transformers: Modding LegalBERT and Longformer