arXiv cs.CL· Lian Yan, Zhenbo Xia, Haotian Wang, Chen Tang, Yi Guan, Jie Liu, Jingchi Jiang·· 12 小时前AI 评分24
CoLink：面向多轮对话建模的全局压缩与知识链接框架
CoLink: Global Compression and Knowledge Linking for Multi-Turn Dialogue Modeling
AI 导读
CoLink 框架通过信息链接与压缩提升 LLM 在领域特定多轮对话中的全局上下文感知能力，先将历史上下文聚类为主题图并与外部知识建立跨语义关联，再结合语义相关性与图结构拓扑检索激活知识。该框架基于信息论推导出提取核心意义信息的优化目标，配合竞争槽机制将全局上下文压缩为紧凑的潜在表示。在公开数据集与自建 MMD 数据集上的实验验证了其在知识链接与意义建模上的有效性和泛化性。
正文
Abstract:Extracting salient information from long-horizon dialogue histories remains a significant challenge for Large Language Models (LLMs) due to the restricted context windows and efficient retrieval challenges. Current methods typically rely on history organization or local token-level optimization, suffering from information loss, retrieval redundancy, and myopic attention. Inspired by the view in cognitive science that dialogue constitutes sense-making and memory association, we propose CoLink, a framework that enhances LLMs' global context awareness in domain-specific multi-turn dialogues through information linking and compression. CoLink first clusters historical contexts into topic graphs to establish cross-semantic associations with external knowledge. Guided by the adaptive control of thought, it then retrieves and activates knowledge via synergizing semantic relevance and graph-structural topology. Crucially, to achieve effective compression of history and knowledge, we ground multi-turn dialogues in information theory and derive a principled optimization objective for extracting the core sense-making information. Coupled with a competitive slot mechanism, CoLink condenses the global context into dedicated memory and knowledge slots, efficiently distilling lengthy, redundant inputs into an extremely compact latent representation. Extensive empirical results on both public and our MMD datasets validate CoLink's superiority and generalizability in knowledge linking and sense modeling.
| Comments: | 7 pages, 15 figures, work in prograss |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.34335 [cs.CL] |
| (or arXiv:2610.34335v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.34335 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Haotian Wang [view email]
[v1]
Sun, 25 Oct 2026 14:51:19 UTC (3,682 KB)
来源：arXiv cs.CL · arxiv.org