Let's Revise Step-by-Step: A Unified Local Search Framework for Code Generation with LLMs

• URL: http://arxiv.org/abs/2508.07434v1
• Date: Sun, 10 Aug 2025 17:11:56 GMT
• Title: Let's Revise Step-by-Step: A Unified Local Search Framework for Code Generation with LLMs
• Authors: Zhiyi Lyu, Jianguo Huang, Yanchen Deng, Steven Hoi, Bo An,
• Abstract summary: We propose a unified local search framework which effectively performs step-by-step code revision.<n>Specifically, ReLoc explores a series of local revisions through four key algorithmic components.<n>We develop a specialized revision reward model that evaluates code quality based on revision distance to produce fine-grained preferences.
• Score: 16.818072348542923
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) with inference-time scaling techniques show promise for code generation, yet face notable efficiency and scalability challenges. Construction-based tree-search methods suffer from rapid growth in tree size, high token consumption, and lack of anytime property. In contrast, improvement-based methods offer better performance but often struggle with uninformative reward signals and inefficient search strategies. In this work, we propose \textbf{ReLoc}, a unified local search framework which effectively performs step-by-step code revision. Specifically, ReLoc explores a series of local revisions through four key algorithmic components: initial code drafting, neighborhood code generation, candidate evaluation, and incumbent code updating, each of which can be instantiated with specific decision rules to realize different local search algorithms such as Hill Climbing (HC) or Genetic Algorithm (GA). Furthermore, we develop a specialized revision reward model that evaluates code quality based on revision distance to produce fine-grained preferences that guide the local search toward more promising candidates. Finally, our extensive experimental results demonstrate that our approach achieves superior performance across diverse code generation tasks, significantly outperforming both construction-based tree search as well as the state-of-the-art improvement-based code generation methods.

Related papers

• MaxCode: A Max-Reward Reinforcement Learning Framework for Automated Code Optimization [44.27213441671799]
  Large Language Models (LLMs) demonstrate strong capabilities in general coding tasks but encounter two key challenges when optimizing code.<n>In this work, we explore inference-time search algorithms that guide the LLM to discover better solutions.<n>Our approach, called MaxCode unifies existing search methods under a max-reward reinforcement learning framework.
  arXiv  Detail & Related papers  (2026-01-09T02:21:28Z)
• ReCode: Improving LLM-based Code Repair with Fine-Grained Retrieval-Augmented Generation [26.836612605244596]
  We propose ReCode, a fine-grained retrieval-augmented in-context learning framework for accurate and efficient code repair.<n>ReCode introduces two key innovations: (1) an algorithm-aware retrieval strategy that narrows the search space using preliminary algorithm type predictions; and (2) a modular dual-encoder architecture that separately processes code and textual inputs.<n> Experimental results on RACodeBench and competitive programming datasets demonstrate that ReCode achieves higher repair accuracy with significantly reduced inference cost.
  arXiv  Detail & Related papers  (2025-09-02T13:58:48Z)
• Alignment with Fill-In-the-Middle for Enhancing Code Generation [56.791415642365415]
  We propose a novel approach that splits code snippets into smaller, granular blocks, creating more diverse DPO pairs from the same test cases.<n>Our approach demonstrates significant improvements in code generation tasks, as validated by experiments on benchmark datasets such as HumanEval (+), MBPP (+), APPS, LiveCodeBench, and BigCodeBench.
  arXiv  Detail & Related papers  (2025-08-27T03:15:53Z)
• LSR-MCTS: Alleviating Long Range Dependency in Code Generation [42.10272627826627]
  Large language models (LLMs) have significantly promoted the development of code generation task.<n>We propose the textbfLSR-MCTS algorithm, which leverages MCTS to determine the code line-by-line and select the optimal path.
  arXiv  Detail & Related papers  (2025-04-10T04:03:25Z)
• CodeXEmbed: A Generalist Embedding Model Family for Multiligual and Multi-task Code Retrieval [103.116634967815]
  We introduce CodeXEmbed, a family of large-scale code embedding models ranging from 400M to 7B parameters.<n>Our novel training pipeline unifies multiple programming languages and transforms various code-related tasks into a common retrieval framework.<n>Our 7B model sets a new state-of-the-art (SOTA) in code retrieval, outperforming the previous leading model, Voyage-Code, by over 20% on CoIR benchmark.
  arXiv  Detail & Related papers  (2024-11-19T16:54:45Z)
• Chain-of-Programming (CoP) : Empowering Large Language Models for Geospatial Code Generation [2.6026969939746705]
  This paper proposes a Chain of Programming framework to decompose the code generation process into five steps. The framework incorporates a shared information pool, knowledge base retrieval, and user feedback mechanisms. It significantly improves the logical clarity, syntactical correctness, and executability of the generated code.
  arXiv  Detail & Related papers  (2024-11-16T09:20:35Z)
• CodeTree: Agent-guided Tree Search for Code Generation with Large Language Models [106.11371409170818]
  Large language models (LLMs) can act as agents with capabilities to self-refine and improve generated code autonomously. We propose CodeTree, a framework for LLM agents to efficiently explore the search space in different stages of the code generation process. Specifically, we adopted a unified tree structure to explicitly explore different coding strategies, generate corresponding coding solutions, and subsequently refine the solutions.
  arXiv  Detail & Related papers  (2024-11-07T00:09:54Z)
• RethinkMCTS: Refining Erroneous Thoughts in Monte Carlo Tree Search for Code Generation [71.88883580383039]
  We propose RethinkMCTS, a framework that explores and refines the reasoning process for code generation.<n>Specifically, we employ MCTS to search for thoughts before code generation and integrate MCTS with a refinement mechanism called rethink.<n>We demonstrate that RethinkMCTS outperforms previous search-based and feedback-enhanced code generation baselines.
  arXiv  Detail & Related papers  (2024-09-15T02:07:28Z)
• DOCE: Finding the Sweet Spot for Execution-Based Code Generation [69.5305729627198]
  We propose a comprehensive framework that includes candidate generation, $n$-best reranking, minimum Bayes risk (MBR) decoding, and self-ging as the core components. Our findings highlight the importance of execution-based methods and the difference gap between execution-based and execution-free methods.
  arXiv  Detail & Related papers  (2024-08-25T07:10:36Z)
• StepCoder: Improve Code Generation with Reinforcement Learning from Compiler Feedback [58.20547418182074]
  We introduce StepCoder, a novel framework for code generation, consisting of two main components. CCCS addresses the exploration challenge by breaking the long sequences code generation task into a Curriculum of Code Completion Subtasks. FGO only optimize the model by masking the unexecuted code segments to provide Fine-Grained Optimization. Our method improves the ability to explore the output space and outperforms state-of-the-art approaches in corresponding benchmarks.
  arXiv  Detail & Related papers  (2024-02-02T13:14:31Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.