Scholarlink

<p>Address parsing seeks to map noisy, abbreviated free-text addresses into standardized hierarchical tuples for large-scale information systems. Existing approaches struggle with semantic and structural ambiguity, hallucination from unconstrained generation, and deployment constraints under privacy and governance requirements. We present AddrKG-LLM, a two-stage framework that combines knowledge-graph (KG)–aware retrieval with schema-restricted large language model (LLM) decoding. First, contrastive learning over multi-view administrative graphs yields node embeddings that retrieve and re-rank a compact Top-K candidate set, thereby bounding the search space while preserving high gold coverage (Recall@K). Second, a candidate-restricted decoder running on-premises produces JSON-compliant outputs, enforcing single-candidate field consistency and alignment with KG priors to improve controllability and policy compliance. Using de-identified real-world records, we evaluate structural consistency via micro-level accuracy (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(A_{micro}\)</EquationSource> </InlineEquation>) and macro-level accuracy (<InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(A_{macro}\)</EquationSource> </InlineEquation>), and assess system properties with Recall@K and latency. Across strong string-matching, sequence-labeling, and generic LLM baselines, AddrKG-LLM delivers consistent gains in <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(A_{micro}\)</EquationSource> </InlineEquation> and <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(A_{macro}\)</EquationSource> </InlineEquation> with a favorable Recall@K. The proposed method consists of three components: (i) multi-view graph aggregation, (ii) a hierarchy-aware self-supervised contrastive objective that derives positives/negatives from administrative relations to align textual and graph embeddings, and (iii) candidate-restricted decoding within the KG-derived Top-K set. Overall, coupling KG-aware retrieval with constrained on-prem LLM decoding yields an accurate, controllable, and deployable solution for noisy-address structuring across domains.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Patient address parsing via KG-aware contrastive learning and constrained on-prem LLM inference

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月9日

查看全文
  • Jinzhe Li,
  • Xin Pan,
  • Yanchao Jia

摘要

Address parsing seeks to map noisy, abbreviated free-text addresses into standardized hierarchical tuples for large-scale information systems. Existing approaches struggle with semantic and structural ambiguity, hallucination from unconstrained generation, and deployment constraints under privacy and governance requirements. We present AddrKG-LLM, a two-stage framework that combines knowledge-graph (KG)–aware retrieval with schema-restricted large language model (LLM) decoding. First, contrastive learning over multi-view administrative graphs yields node embeddings that retrieve and re-rank a compact Top-K candidate set, thereby bounding the search space while preserving high gold coverage (Recall@K). Second, a candidate-restricted decoder running on-premises produces JSON-compliant outputs, enforcing single-candidate field consistency and alignment with KG priors to improve controllability and policy compliance. Using de-identified real-world records, we evaluate structural consistency via micro-level accuracy ( \(A_{micro}\) ) and macro-level accuracy ( \(A_{macro}\) ), and assess system properties with Recall@K and latency. Across strong string-matching, sequence-labeling, and generic LLM baselines, AddrKG-LLM delivers consistent gains in \(A_{micro}\) and \(A_{macro}\) with a favorable Recall@K. The proposed method consists of three components: (i) multi-view graph aggregation, (ii) a hierarchy-aware self-supervised contrastive objective that derives positives/negatives from administrative relations to align textual and graph embeddings, and (iii) candidate-restricted decoding within the KG-derived Top-K set. Overall, coupling KG-aware retrieval with constrained on-prem LLM decoding yields an accurate, controllable, and deployable solution for noisy-address structuring across domains.