SeDev: Structured Semantic Exploration for LLM-Driven Code Generation

Large Language Models (LLMs) have shown remarkable capabilities in automating code generation. Recent approaches that incorporate feedback refinement mechanisms into the generation process have further enhanced software generation quality. However, these methods can be characterized as single-path approaches, which suffer from insufficient exploration of the vast solution space, often causing even the most powerful models to get stuck in local optima and struggle to generate the desired software. Some other works use Monte Carlo Tree Search (MCTS) to explore multiple paths for finding the best solution; yet, MCTS can be extremely inefficient in practice. To this end, we propose SeDev, a novel LLM-driven code generation framework that efficiently finds highquality solutions in only a few iterations. The core idea of SeDev is to gradually explore semantically adjacent solutions through structured prompt guidance and feedback on previous trials, while using unit tests to evaluate the quality of exploration. To distill the exploration experience, SeDev incorporates a feedback synthesis module that translates unit test results within exploration into comprehensive suggestions. We construct a challenging software benchmark FSD-Bench++, along with two open datasets to evaluate. Experimental results show that SeDev outperforms baselines while maintaining reasonable time and computational costs. Code is available here. Plan Code Architecture Problem: Potential issues with note retrieval when titles are identical. Guidance: Update the get_note, view_note, and note management logic. Problem: Functions referenced in the view_note.html are not implemented. Guidance: Tasks like implementing note editing and deletion should be prioritized immediately after the core functionalities.