Citation-Grounded Code Comprehension: Preventing LLM Hallucination Through Hybrid Retrieval and Graph-Augmented Context
This addresses the critical barrier of unreliable developer assistance due to factually incorrect citations in code comprehension systems, though it appears incremental as it builds on existing retrieval and graph techniques.
The paper tackles the problem of LLM hallucination in code comprehension by developing a hybrid retrieval system combining BM25 sparse matching, BGE dense embeddings, and Neo4j graph expansion, achieving 92% citation accuracy with zero hallucinations and outperforming baselines by 14-18 percentage points.
Large language models have become essential tools for code comprehension, enabling developers to query unfamiliar codebases through natural language interfaces. However, LLM hallucination, generating plausible but factually incorrect citations to source code, remains a critical barrier to reliable developer assistance. This paper addresses the challenges of achieving verifiable, citation grounded code comprehension through hybrid retrieval and lightweight structural reasoning. Our work is grounded in systematic evaluation across 30 Python repositories with 180 developer queries, comparing retrieval modalities, graph expansion strategies, and citation verification mechanisms. We find that challenges of citation accuracy arise from the interplay between sparse lexical matching, dense semantic similarity, and cross file architectural dependencies. Among these, cross file evidence discovery is the largest contributor to citation completeness, but it is largely overlooked because existing systems rely on pure textual similarity without leveraging code structure. We advocate for citation grounded generation as an architectural principle for code comprehension systems and demonstrate this need by achieving 92 percent citation accuracy with zero hallucinations. Specifically, we develop a hybrid retrieval system combining BM25 sparse matching, BGE dense embeddings, and Neo4j graph expansion via import relationships, which outperforms single mode baselines by 14 to 18 percentage points while discovering cross file evidence missed by pure text similarity in 62 percent of architectural queries.