用大模型解决多语言指代消解，效果超越传统方法。
CorefInst: Leveraging LLMs for Multilingual Coreference Resolution
- 用指令微调的Decoder-only大模型处理显性和隐性指代
- 在CorefUD v1.2上平均性能领先现有模型2个百分点
- 适用于多语言场景，无需为每语言定制架构
指代消解（CR）是自然语言理解中的关键任务，常受特定任务架构和编码器模型限制，需大量训练且难以适应新场景。本文首次提出利用Decoder-only大模型进行多语言指代消解的方法，可同时处理显性和零指代。通过五种不同指令集与可控推理方式，在Llama 3.1、Gemma 2和Mistral 0.3三类大模型上评估该方法。结果表明，经适当指令微调的大模型可超越现有专用架构。其中，全微调的Llama 3.1在CorefUD v1.2数据集所有语言上平均性能超过领先模型Corpipe 24单阶段变体2个百分点。
原文摘要 · Abstract (English)
Coreference Resolution (CR) is a crucial yet challenging task in natural language understanding, often constrained by task-specific architectures and encoder-based language models that demand extensive training and lack adaptability. This study introduces the first multilingual CR methodology which leverages decoder-only LLMs to handle both overt and zero mentions. The article explores how to model the CR task for LLMs via five different instruction sets using a controlled inference method. The approach is evaluated across three LLMs; Llama 3.1, Gemma 2, and Mistral 0.3. The results indicate that LLMs, when instruction-tuned with a suitable instruction set, can surpass state-of-the-art task-specific architectures. Specifically, our best model, a fully fine-tuned Llama 3.1 for multilingual CR, outperforms the leading multilingual CR model (i.e., Corpipe 24 single stage variant) by 2 pp on average across all languages in the CorefUD v1.2 dataset collection.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。