Papers by Mohamed Hesham Elganayni

Copied to clipboard

Challenge:	Existing approaches to training language models for each jurisdiction fail to leverage common legal principles beneficial for low-resource settings or risk negative interference from conflicting jurisdictional interpretations.
Approach:	They propose a parameter-efficient framework that derives hierarchical relationships across jurisdictions and progressively inserts adapter modules across model layers based on jurisdictional similarity.
Outcome:	The proposed framework outperforms fully shared and jurisdiction-specific models on two legal language modeling benchmarks.