InfCode-C++: Intent-Guided Semantic Retrieval and AST-Structured Search for C++ Issue Resolution

2025-11-19

AI Summary

Existing LLM-based agents exhibit significantly inferior performance in repairing C++ projects compared to Python, primarily due to their reliance on lexical retrieval and shallow code navigation—approaches ill-suited for C++’s complexities, including overloaded identifiers, nested namespaces, template instantiations, and intricate control flow. Method: We propose the first end-to-end automated repair system tailored for C++. It integrates intent-guided semantic retrieval with AST-structured querying to construct language-aware, precise contextual representations and localize defects accurately. Our approach explicitly models C++’s static typing, multi-namespace scoping, and template semantics through fault-reproduction analysis and collaborative LLM reasoning. Contribution/Results: Evaluated on MultiSWE-bench-CPP, our system achieves a 25.58% solution rate—outperforming the strongest baseline by 10.85 percentage points and more than doubling the performance of MSWE-agent.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Multiagent Planning

Application Category

• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Large language model (LLM) agents have recently shown strong performance on repository-level issue resolution, but existing systems are almost exclusively designed for Python and rely heavily on lexical retrieval and shallow code navigation. These approaches transfer poorly to C++ projects, where overloaded identifiers, nested namespaces, template instantiations, and deep control-flow structures make context retrieval and fault localization substantially more difficult. As a result, state-of-the-art Python-oriented agents show a drastic performance drop on the C++ subset of MultiSWE-bench. We introduce INFCODE-C++, the first C++-aware autonomous system for end-to-end issue resolution. The system combines two complementary retrieval mechanisms -- semantic code-intent retrieval and deterministic AST-structured querying -- to construct accurate, language-aware context for repair.These components enable precise localization and robust patch synthesis in large, statically typed C++ repositories. Evaluated on the exttt{MultiSWE-bench-CPP} benchmark, INFCODE-C++ achieves a resolution rate of 25.58%, outperforming the strongest prior agent by 10.85 percentage points and more than doubling the performance of MSWE-agent. Ablation and behavioral studies further demonstrate the critical role of semantic retrieval, structural analysis, and accurate reproduction in C++ issue resolution. INFCODE-C++ highlights the need for language-aware reasoning in multi-language software agents and establishes a foundation for future research on scalable, LLM-driven repair for complex, statically typed ecosystems.

Problem

• Addressing poor performance of Python-oriented LLM agents on C++ issue resolution
• Overcoming challenges with overloaded identifiers and complex C++ code structures
• Improving context retrieval and fault localization in statically typed C++ repositories

Innovation

• Intent-guided semantic retrieval for C++ code
• AST-structured search for precise fault localization
• Combining semantic and structural analysis for repair