arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.05400cs.CL
收藏

基于低参数LLM的探索-分析-消歧推理框架用于词义消歧

An Exploration-Analysis-Disambiguation Reasoning Framework for Word Sense Disambiguation with Low-Parameter LLMs

Deshan Sumanathilaka, Nicholas Micallef, Julian Hough

AI总结:

本研究提出基于低参数LLM的探索-分析-消歧推理框架,通过推理驱动的微调策略,在零样本设置中实现与GPT-4-Turbo相当的词义消歧性能,并展示了对未见词义的稳健泛化能力。

摘要:

词义消歧(WSD)仍然是自然语言处理(NLP)中的关键挑战,尤其是在处理罕见或领域特定的词义时,这些词义往往被误判。尽管现代高参数大型语言模型(LLMs)如GPT-4-Turbo在WSD性能上表现领先,但其计算和能耗需求限制了可扩展性。本研究探讨低参数LLM(<4B参数)是否能通过强调推理驱动的词义识别的微调策略实现类似结果。使用增强的FEWS数据集,该数据集包含半自动化的、富含推理的注释,我们对八个小型开源LLM(如Gemma和Qwen)进行了微调。我们的结果表明,在零样本设置中,基于推理链(CoT)的推理结合邻近词分析可达到与GPT-4-Turbo相当的性能。重要的是,Gemma-3-4B和Qwen-3-4B模型在FEWS上始终优于所有中等参数基线和最先进模型,具有对未见词义的稳健泛化能力。此外,对未见“Fool Me If You Can”数据集的评估证实了在无任务特定微调的情况下强大的跨领域适应性。本工作表明,通过精心设计的以推理为中心的微调,低参数LLM可以在大幅降低计算和能耗需求的同时实现准确的WSD。

英文摘要:

Word Sense Disambiguation (WSD) remains a key challenge in Natural Language Processing (NLP), especially when dealing with rare or domain-specific senses that are often misinterpreted. While modern high-parameter Large Language Models (LLMs) such as GPT-4-Turbo have shown state-of-the-art WSD performance, their computational and energy demands limit scalability. This study investigates whether low-parameter LLMs (<4B parameters) can achieve comparable results through fine-tuning strategies that emphasize reasoning-driven sense identification. Using the FEWS dataset augmented with semi-automated, rationale-rich annotations, we fine-tune eight small-scale open-source LLMs (e.g. Gemma and Qwen). Our results reveal that Chain-of-Thought (CoT)-based reasoning combined with neighbour-word analysis achieves performance comparable to GPT-4-Turbo in zero-shot settings. Importantly, Gemma-3-4B and Qwen-3-4B models consistently outperform all medium-parameter baselines and state-of-the-art models on FEWS, with robust generalization to unseen senses. Furthermore, evaluation on the unseen "Fool Me If You Can'' dataset confirms strong cross-domain adaptability without task-specific fine-tuning. This work demonstrates that with carefully crafted reasoning-centric fine-tuning, low-parameter LLMs can deliver accurate WSD while substantially reducing computational and energy demands.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑