Improving Lexical Difficulty Prediction with Context-Aligned Contrastive Learning and Ridge Ensembling

TL;DR

This paper introduces a novel contrastive learning approach combined with ridge ensembling to improve cross-lingual lexical difficulty prediction, capturing ordinal structure and reducing bias.

Contribution

It proposes Context-Aligned Contrastive Regression integrating contrastive objectives with ensemble methods for better lexical difficulty modeling across languages.

Findings

Contrastive objectives enhance cross-lingual representation alignment.

Learned representations effectively capture ordinal difficulty structure.

Ensemble reduces systematic biases, improving stability across difficulty levels.

Abstract

Lexical difficulty prediction is a fundamental problem in language learning and readability assessment, requiring models to estimate word difficulty across different first-language (L1) backgrounds. However, existing approaches rely on regression-only training with scalar supervision, which does not explicitly structure the representation space, limiting their ability to capture cross-lingual alignment and ordinal difficulty. To mitigate these issues, we propose Context-Aligned Contrastive Regression, which integrates Ridge regression ensemble with two complementary objectives, i.e., Cross-View Context and Ordinal Soft Contrastive Learning. Experiments on three L1 datasets show that (i) contrastive objectives improve cross-lingual representation alignment while preserving language-specific nuances, (ii) the learned representations capture the ordinal structure of lexical difficulty, and (iii) the ensemble effectively mitigates systematic biases of individual models, leading to more stable performance across difficulty levels.