Unveiling the semantic potential of BERT for unsupervised Chinese WSD via non-linear semantic sharpening

Use BiLSTM with Attention Mechanism to Optimize the Accuracy of Word Meaning Correspondence in Technical Texts

A robust WSD model that integrates a bidirectional long shortterm memory network (BiLSTM) with an attention mechanism, specifically designed for Chinese patent texts is proposed, providing reliable support for knowledge mining and intelligent text processing in technical domains.

NRD: A Hybrid Disentanglement Framework for Mitigating Interference in Multilingual Machine Translation

Uncovering lexical semantic variation in Mandarin varieties based on word embeddings: the case of Hong Kong written Chinese

Abstract Identifying lexical semantic differences across closely related language varieties remains methodologically challenging, as traditional approaches often rely on intuition, manual inspection, or researcher-constructed word lists with limited coverage. This study proposes a computational framework that uses word...

Semantic Parsing for Evaluating Large Language Models: Separating Linguistic Abilities with YARN

A layer-wise analysis indicates that surface-level features such as temporality and negation are captured more reliably than deeper semantic phenomena like quantification in large language models, highlighting the limited capacity of current LLMs to generate fully formal meaning representations.

WiC is Not WSD: A Study on LLMs and Lexical Ambiguity Resolution

It is found that providing candidate senses, similar to what is done in traditional WSD, improves WiC performance in all settings and helps models make more consistent and targeted judgements.

Linguistic Features for Interpretable Textual Entailment

SLITE is presented, an explainable hybrid model for Recognizing Textual Entailment that integrates two complementary layers of semantic analysis: a structural-relational layer, based on semantic compatibility and incompatibility between compositional entities, and a distributional-informational layer, based on structur...