通过分层协作的LoRA解决对话上下文与提示语不匹配问题，提升零样本对话状态追踪性能。
HiCoLoRA: Addressing Context-Prompt Misalignment via Hierarchical Collaborative LoRA for Zero-Shot DST
- 分层LoRA架构实现动态层级处理，结合低层启发式分组与高层全连接交互
- 在MultiWOZ和SGD数据集上达到当前最佳零样本槽位识别效果
- 适合需要跨领域泛化且无标注数据的新任务对话系统研究者
零样本对话状态追踪（zs-DST）对于实现任务导向对话系统在新领域上的泛化至关重要，无需昂贵的数据标注。核心挑战在于动态对话上下文与静态提示之间的语义不匹配，导致跨层协作僵化、领域干扰和灾难性遗忘。为此，我们提出分层协同低秩适配（HiCoLoRA），通过强化提示对齐来提升零样本槽位推断能力。该框架采用分层LoRA结构实现动态层间处理（结合低层启发式分组与高层全交互），引入谱联合域-槽聚类以识别可迁移关联（输入自适应线性融合机制），并使用语义增强的SVD初始化（SemSVD-Init）以保留预训练知识。在多领域数据集MultiWOZ和SGD上的实验表明，HiCoLoRA优于现有基线，在zs-DST任务中达到当前最优表现。代码已开源：https://github.com/carsonz/HiCoLoRA。
原文摘要 · Abstract (English)
Zero-shot Dialog State Tracking (zs-DST) is essential for enabling Task-Oriented Dialog Systems (TODs) to generalize to new domains without costly data annotation. A central challenge lies in the semantic misalignment between dynamic dialog contexts and static prompts, leading to inflexible cross-layer coordination, domain interference, and catastrophic forgetting. To tackle this, we propose Hierarchical Collaborative Low-Rank Adaptation (HiCoLoRA), a framework that enhances zero-shot slot inference through robust prompt alignment. It features a hierarchical LoRA architecture for dynamic layer-specific processing (combining lower-layer heuristic grouping and higher-layer full interaction), integrates Spectral Joint Domain-Slot Clustering to identify transferable associations (feeding an Adaptive Linear Fusion Mechanism), and employs Semantic-Enhanced SVD Initialization (SemSVD-Init) to preserve pre-trained knowledge. Experiments on multi-domain datasets MultiWOZ and SGD show that HiCoLoRA outperforms baselines, achieving SOTA in zs-DST. Code is available at https://github.com/carsonz/HiCoLoRA.
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。