GRACE：基于层次代码融合的数据驱动图结构代码补全

LLM 在 localized code completion 中表现出色，但由于 limited context windows 和 complex semantic and structural dependencies across codebases， 在 repository-level tasks 中 面临挑战。虽然 Retrieval-Augmented Generation (RAG) 通过检索 relevant code snippets 来缓解 context 稀缺，但 current approaches 面临 significant limitations。 They overly rely on textual similarity for retrieval, neglecting structural relationships such as call chains and inheritance hierarchies, and lose critical structural information by naively concatenating retrieved snippets into text sequences for LLM input。 To address these shortcomings, GRACE constructs a multi-level, multi-semantic code graph that unifies file structures, abstract syntax trees, function call graphs, class hierarchies, and data flow graphs to capture both static and dynamic code semantics。 For retrieval, GRACE employs a Hybrid Graph Retriever that integrates graph neural network-based structural similarity with textual retrieval, refined by a graph attention network-based re-ranker to prioritize topologically relevant subgraphs。 To enhance context, GRACE introduces a structural fusion mechanism that merges retrieved subgraphs with the local code context and preserves essential dependencies like function calls and inheritance。 Extensive experiments on public repository-level benchmarks demonstrate that GRACE significantly outperforms state-of-the-art methods across all metrics。 Using DeepSeek-V3 as the backbone LLM, GRACE surpasses the strongest graph-based RAG baselines by 8.19% EM and 7.51% ES points on every dataset。 The code is available at https://anonymous.4open.science/r/grace_icse-C3D5。