代码大模型 / AI 编程

Incoherence as Oracle-less Measure of Error in LLM-Based Code Generation

不一致性作为无 oracle 的错误度量方法在基于 LLM 的代码生成中的应用

Thomas Valentin, Ardi Madadi, Gaetano Sapia, Marcel Böhme

专题命中 代码生成 ：code generation(title)；分类 cs.SE、cs.AI、cs.LG

AI总结 本文提出了一种无 oracle 的错误度量方法，用于评估基于LLM的代码生成的正确性，通过不一致性度量有效识别错误程序并替代传统 oracle 评估。

Comments Accepted at AAAI'26 (extended version). 8 pages + refs and appendix

Journal ref 40th Annual AAAI Conference on Artificial Intelligence (AAAI), 2026

AutoFSM: A Multi-agent Framework for FSM Code Generation with IR and SystemC-Based Testing

AutoFSM：一种基于IR和SystemC的多智能体FSM代码生成框架

Qiuming Luo, Yanming Lei, Kunzhong Wu, Yixuan Cao, Chengjian Liu

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

AI总结 AutoFSM通过引入结构化中间表示和SystemC建模，提升了FSM代码生成的可靠性与可扩展性，实验表明其在测试基准中表现优于现有框架。

Comments This version corrects a typo in the section title ("Intruction" -> "Introduction") that appears in the published version

Understanding Chain-of-Thought Effectiveness in Code Generation: An Empirical and Information-Theoretic Analysis

理解链式思维在代码生成中的有效性：一项经验性与信息论分析

Naizhu Jin, Zhong Li, Guang Yang, Tian Zhang, Qingkai Zeng

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

AI总结 本研究通过经验分析和信息论方法，揭示了链式思维在代码生成中有效性的影响因素，发现结构化CoT在提升生成质量方面表现更优，且其效果受语言系统和模型容量影响。

A Hybrid Approach for EMF Code Generation:Code Templates Meet Large Language Models

一种混合方法用于EMF代码生成：代码模板与大型语言模型的结合

Xiao He, Ru Chen, Zeqing Zhang, Yanling Wang, Qiuyan Dong

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

AI总结 本文提出iEcoreGen，结合EMF和LLM，通过模板生成初始代码并利用LLM完善方法，提升代码生成效率与准确性。

When Forgetting Builds Reliability: LLM Unlearning for Reliable Hardware Code Generation

当遗忘构建可靠性：用于可靠硬件代码生成的LLM去学习

Yiwen Liang, Qiufeng Li, Shikai Wang, Weidong Cao

机构 * Department of ECE, The George Washington University（电子工程系，乔治·华盛顿大学） 纠错

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.LG

AI总结 本文提出了一种针对LLM硬件代码生成的去学习框架，通过语法保持策略和细粒度损失机制有效去除不良知识，提升代码生成的可靠性。

Completion by Comprehension: Guiding Code Generation with Multi-Granularity Understanding

通过理解完成：利用多粒度理解引导代码生成

Xinkui Zhao, Rongkai Liu, Yifan Zhang, Chen Zhi, Lufei Zhang, Guanjie Cheng, Yueshen Xu, Shuiguang Deng, Jianwei Yin

专题命中 代码生成 ：code generation(title)；repository(abstract)；分类 cs.SE

AI总结 CoCo通过多粒度理解引导代码生成，利用结构化上下文提升代码补全质量，实验表明其在多个基准测试中表现优异。

PennyCoder: Efficient Domain-Specific LLMs for PennyLane-Based Quantum Code Generation

PennyCoder: 用于PennyLane量子代码生成的高效领域特定LLM

Abdul Basit, Minghao Shao, Muhammad Haider Asif, Nouhaila Innan, Muhammad Kashif, Alberto Marchisio, Muhammad Shafique

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.AI

AI总结 PennyCoder是一种针对PennyLane量子编程的高效领域特定LLM，通过本地部署和低秩适应技术提升量子代码生成的准确性和效率。

Comments 6 pages, 5 figures, 3 tables, paper accepted to QCE 2025

Journal ref 2025 IEEE International Conference on Quantum Computing and Engineering (QCE), Albuquerque, NM, USA

GeoJSON Agents:A Multi-Agent LLM Architecture for Geospatial Analysis-Function Calling vs Code Generation

GeoJSON代理：一种用于地理空间分析的多智能体LLM架构——功能调用与代码生成

Qianqian Luo, Qingming Lin, Liuchang Xu, Sensen Wu, Ruichen Mao, Chao Wang, Hailin Feng, Bo Huang, Zhenhong Du

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

AI总结 GeoJSON代理通过功能调用与代码生成技术，提升地理空间分析任务的准确性和稳定性，展现多智能体LLM在复杂任务中的优势。

Accelerated Execution of Bayesian Neural Networks using a Single Probabilistic Forward Pass and Code Generation

利用单次概率前向传递和代码生成加速贝叶斯神经网络

Bernhard Klein, Falk Selker, Hendrik Borras, Sophie Steger, Franz Pernkopf, Holger Fröning

机构 * Heidelberg University（海德堡大学） ； Graz University of Technology（格拉茨技术大学） 纠错

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.LG

AI总结 本文提出一种基于概率前向传递和代码生成的方法，显著提升贝叶斯神经网络在嵌入式系统中的计算效率和部署性能。

Chart2Code-MoLA: Efficient Multi-Modal Code Generation via Adaptive Expert Routing

Chart2Code-MoLA: 通过自适应专家路由实现高效的多模态代码生成

Yifei Wang, Jacky Keung, Zhenyu Mao, Jingyu Zhang, Yuchen Cao

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

AI总结 Chart2Code-MoLA通过自适应专家路由实现高效多模态代码生成，提升生成准确性、降低内存消耗并加速收敛。

Secure-Instruct: An Automated Pipeline for Synthesizing Instruction-Tuning Datasets Using LLMs for Secure Code Generation

Secure-Instruct: 一种利用LLM合成指令微调数据集的自动化流程，用于安全代码生成

Junjie Li, Fazle Rabbi, Bo Yang, Song Wang, Jinqiu Yang

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

AI总结 Secure-Instruct通过自动化合成安全代码示例和指令微调LLM，提升代码生成的安全性和功能性正确性。

GenSIaC: Toward Security-Aware Infrastructure-as-Code Generation with Large Language Models

Yikun Li, Matteo Grella, Daniel Nahmias, Gal Engelberg, Dan Klein, Giancarlo Guizzardi, Thijs van Ede, Andrea Continella

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

Evaluating Software Process Models for Multi-Agent Class-Level Code Generation

Wasique Islam Shafin, Md Nakhla Rafi, Zhenhao Li, Tse-Hsun Chen

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

Retriv at BLP-2025 Task 2: Test-Driven Feedback-Guided Framework for Bangla-to-Python Code Generation

K M Nafi Asib, Sourav Saha, Mohammed Moshiul Hoque

机构 * Department of Computer Science and Engineering（计算机科学与工程系） 纠错

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.CL

Comments 8 pages, 1 figure, experimental scripts publicly available at https://github.com/NafiAsib/Retriv-BLP25-Task-2

DRIVE: Data Curation Best Practices for Reinforcement Learning with Verifiable Reward in Competitive Code Generation

Speed Zhu, Jianwei Cai, Guang Chen, Lulu Wu, Saiyong Yang, Wiggin Zhou

机构 * Hunyuan Team, Tencent（腾讯 Hunyuan 团队） 纠错

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.LG

Comments 15 pages, 8 figures

PSD2Code: Automated Front-End Code Generation from Design Files via Multimodal Large Language Models

Yongxi Chen, Lei Chen

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

Towards Realistic Project-Level Code Generation via Multi-Agent Collaboration and Semantic Architecture Modeling

Qianhui Zhao, Li Zhang, Fang Liu, Junhang Cheng, Chengru Wu, Junchen Ai, Qiaoyuanhe Meng, Lichen Zhang, Xiaoli Lian, Shubin Song, Yuanping Guo

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

Execution Guided Line-by-Line Code Generation

Boaz Lavon, Shahar Katz, Lior Wolf

机构 * Blavatnik School of Computer Science and AI, Tel Aviv University（巴尔-以色列计算机科学与人工智能学院，特拉维夫大学） 纠错

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.LG

Comments Accepted to NeurIPS 2026

SEER: Enhancing Chain-of-Thought Code Generation through Self-Exploring Deep Reasoning

Shuzheng Gao, Chaozheng Wang, Cuiyun Gao, Michael R. Lyu

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

Comments The paper was completed in Feb. 2025, submitted to ICSE 2026 in Mar. 2025, received a major revision in Jun. 2025, and was finally accepted in Oct. 2025

QuanBench: Benchmarking Quantum Code Generation with Large Language Models

Xiaoyu Guo, Minggu Wang, Jianjun Zhao

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

Comments This paper was accepted by ASE2025

Online automatic code generation for robot swarms: LLMs and self-organizing hierarchy

Weixu Zhu, Marco Dorigo, Mary Katherine Heinrich

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.AI

Comments This abstract was accepted to and presented at the "Multi-Agent Cooperative Systems and Swarm Robotics in the Era of Generative AI" (MACRAI) workshop at the 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2025)

Self-planning Code Generation with Large Language Models

Xue Jiang, Yihong Dong, Lecheng Wang, Zheng Fang, Qiwei Shang, Ge Li, Zhi Jin, Wenpin Jiao

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

Comments Accepted by TOSEM

A11YN: aligning LLMs for accessible web UI code generation

Janghan Yoon, Jaegwan Cho, Junhyeok Kim, Jiwan Chung, Jaehyun Jeon, Youngjae Yu

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

A Matter of Representation: Towards Graph-Based Abstract Code Generation

Nyx Iskandar, Hisham Bedri, Andy Tsen

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.CL

GeoJSEval: An Automated Evaluation Framework for Large Language Models on JavaScript-Based Geospatial Computation and Visualization Code Generation

Guanyu Chen, Haoyue Jiao, Shuyang Hou, Ziqi Liu, Lutong Xie, Shaowen Wu, Huayi Wu, Xuefeng Guan, Zhipeng Gui

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

Journal ref 2025

Toward Green Code: Prompting Small Language Models for Energy-Efficient Code Generation

Humza Ashraf, Syed Muhammad Danish, Shadikur Rahman, Zeeshan Sattar

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

DeepV: A Model-Agnostic Retrieval-Augmented Framework for Verilog Code Generation with a High-Quality Knowledge Base

Zahin Ibnat, Paul E. Calzada, Rasin Mohammed Ihtemam, Sujan Kumar Saha, Jingbo Zhou, Farimah Farahmandi, Mark Tehranipoor

机构 * University of Florida（佛罗里达大学） 纠错

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.AI

Comments 22 pages, 6 figures

GramTrans: A Better Code Representation Approach in Code Generation

Zhao Zhang, Qingyuan Liang, Zeyu Sun, Yizhou Chen, Guoqing Wang, Yican Sun, Lu Zhang, Ge Li, Yingfei Xiong

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

Agent-based code generation for the Gammapy framework

Dmitriy Kostunin, Vladimir Sotnikov, Sergo Golovachev, Abhay Mehta, Tim Lukas Holch, Elisa Jones

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.SE

Comments ICRC2025 proceedings PoS(ICRC2025)753

Memorize or Generalize? Evaluating LLM Code Generation with Code Rewriting

Lizhe Zhang, Wentao Chen, Li Zhong, Letian Peng, Zilong Wang, Jingbo Shang

机构 * University of California, San Diego（加州大学圣地亚哥分校） 纠错

专题命中 代码生成 ：code generation(title,abstract)；分类 cs.AI