LANE: Lexical Adversarial Negative Examples for Word Sense Disambiguation

2025-11-14

AI Summary

Neural language models often over-rely on global sentence-level semantics in fine-grained word sense disambiguation (WSD), neglecting local semantic nuances of target words. To address this, we propose a model-agnostic adversarial contrastive learning framework that constructs lexical-level fine-grained negative samples via selective word substitution, explicitly guiding models to attend to context-sensitive meanings of target words and enhancing discriminability among polysemous representations within the same sentence. Unlike conventional contrastive learning approaches that operate on global sentence embeddings, our method shifts focus to localized, context-dependent lexical semantics, thereby improving sensitivity to subtle semantic distinctions. Empirical results demonstrate significant improvements over standard contrastive learning baselines on both WSD and lexical semantic change detection tasks. Qualitative analysis further confirms its effectiveness in capturing fine-grained sense distinctions in complex, contextually rich settings.

Technology Category

• Natural Language Processing: Lexical Semantics and Morphology
• Machine Learning: Adversarial Learning & Robustness
• Multiagent Systems: Adversarial Agents

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Fine-grained word meaning resolution remains a critical challenge for neural language models (NLMs) as they often overfit to global sentence representations, failing to capture local semantic details. We propose a novel adversarial training strategy, called LANE, to address this limitation by deliberately shifting the model's learning focus to the target word. This method generates challenging negative training examples through the selective marking of alternate words in the training set. The goal is to force the model to create a greater separability between same sentences with different marked words. Experimental results on lexical semantic change detection and word sense disambiguation benchmarks demonstrate that our approach yields more discriminative word representations, improving performance over standard contrastive learning baselines. We further provide qualitative analyses showing that the proposed negatives lead to representations that better capture subtle meaning differences even in challenging environments. Our method is model-agnostic and can be integrated into existing representation learning frameworks.

Problem

• Addresses neural models' overfitting to global sentence representations
• Improves fine-grained word meaning resolution through adversarial training
• Enhances discrimination between subtle meaning differences in words

Innovation

• Adversarial training shifts focus to target words
• Generates negatives by marking alternate words
• Improves discriminative word representation separability