LLM Code Generation

LLM: Large Language Model

This field studies how large language models generate, edit, translate, and test programs from natural-language instructions, examples, or existing code. Common approaches include prompting and retrieval, fine-tuning and reinforcement learning, constrained decoding, and agentic loops that use execution, tests, or static analysis to refine outputs. Evaluations range from HumanEval, MBPP, and LiveCodeBench to repository-level and domain-specific tasks, increasingly measuring security, efficiency, and adherence to requirements alongside whether code runs. Models have improved on standard benchmarks, but reliability on unfamiliar APIs, complex projects, specialized languages, and implicit requirements remains a challenge, as do trade-offs among correctness, safety, cost, and maintainability.

Latest papers

314 papers23 papers in the last four weeks, down 12% on the four weeks before. 0.1% of all new papers.

ACE: Self-Evolving LLM Coding Framework via Adversarial Unit Test Generation and Preference Optimization
Apr 17, 2026Yixu Huang, Xinglei Yu, Zhongyu WeiExecution-Guided Code Generation Language Model Self-Improvement
School of Data Science Fudan University Shanghai, China
HYPERHEURIST: A Simulated Annealing-Based Control Framework for LLM-Driven Code Generation in Optimized Hardware Design
Apr 17, 2026Shiva Ahir, Prajna Bhat, Alex DoboliLarge Language Model-Guided Optimization Code Generation
Department of ECE
QuantCode-Bench: A Benchmark for Evaluating the Ability of Large Language Models to Generate Executable Algorithmic Trading Strategies
Apr 16, 2026Alexey Khoroshilov, Alexey Chernysh, Orkhan Ekhtibarov +2Agentic Code Generation Code Generation Evaluation
Lime
Vibe-Coding: Feedback-Based Automated Verification with no Human Code Inspection, a Feasibility Study
Apr 16, 2026Michal Töpfer, František Plášil, Tomáš Bureš +1LLM Self-Correction Formal Verification
Charles University Prague, Czech Republic
StoryCoder: Narrative Reformulation for Structured Reasoning in LLM Code Generation
Apr 16, 2026Geonhui Jang, Dongyoon Han, YoungJoon YooLLM Prompting LLM Code Generation
Dept. of Artificial Intelligence, Chung-Ang University · NAVER AI Lab
CoDe-R: Refining Decompiler Output with LLMs via Rationale Guidance and Adaptive Inference
Apr 14, 2026Qiang Zhang, Zhongnian LiAdaptive Inference Software Reverse Engineering
School of Computer Science and Technology / School of Artificial Intelligence, China University of Mining and Technology, Xuzhou, China · Mine Digitization Engineering Research Center of the Ministry of Education, China University of Mining and Technology, Xuzhou, China
Is Vibe Coding the Future? An Empirical Assessment of LLM Generated Codes for Construction Safety
Apr 14, 2026S M Jamil UddinConstruction Safety LLM Code Generation
Assistant Professor, Department of Construction Management, Colorado State University
Vibe-driven model-based engineering
Apr 12, 2026Jordi CabotModel-Driven Engineering Software Engineering
Luxembourg Institute of Science and Technology, Esch-sur-Alzette, Luxembourg · University of Luxembourg, Esch-sur-Alzette, Luxembourg
Structure-Grounded Knowledge Retrieval via Code Dependencies for Multi-Step Data Reasoning
Apr 12, 2026Xinyi HuangKnowledge Augmentation for Language Models Retrieval-Augmented Generation
Simon Fraser University
How Many Tries Does It Take? Iterative Self-Repair in LLM Code Generation Across Model Scales and Benchmarks
Apr 12, 2026Johin Johny ArimburLLM Self-Correction LLM Code Generation
Independent Researcher
DeepGuard: Secure Code Generation via Multi-Layer Semantic Aggregation
Apr 10, 2026Li Huang, Zhongxin Liu, Yifan Wu +6Secure Code Generation LLM Code Generation
Chongqing University · The State Key Laboratory of Blockchain and Data Security, Zhejiang University · Peking University
Beyond the Autoregressive Horizon: A Comprehensive Survey of Diffusion Models, World Modelling, and State Space Models for Code
Apr 9, 2026Kishan Maharaj, Ashita Saxena, Srikanth TamilselvamCode World Models LLM Code Generation
IBM
An Empirical Analysis of Static Analysis Methods for Detection and Mitigation of Code Library Hallucinations
Apr 9, 2026Clarissa Miranda-Pena, Andrew Reeson, Cécile Paris +2LLM Hallucination Mitigation Hallucination Detection
The University of Sydney · CSIRO’s Data61
Bridging Natural Language and Interactive What-If Interfaces via LLM-Generated Declarative Specifications
Apr 8, 2026Sneha Gathani, Sirui Zeng, Diya Patel +5LLM Code Generation Text Generation
University of Maryland, College Park College Park, Maryland, USA · Adobe Research San Jose, California, USA · Microsoft Research Seattle, Washington, USA +2
Evaluating LLM-Based 0-to-1 Software Generation in End-to-End CLI Tool Scenarios
Apr 8, 2026Ruida Hu, Xinchen Wang, Chao Peng +2Code Generation Code Generation Evaluation
Harbin Institute of Technology, Shenzhen, China · Independent Researcher, China · Singapore Management University, Singapore
Conv-to-Bench: Evaluating Language Models Via User-Assistant Dialogues In Code Tasks
Apr 7, 2026Victor M. dos Santos, Andre C. Castro, Samuel L. de S. Toledo +5LLM Evaluation Synthetic Benchmark Generation
Institute of Mathematics and Computer Science, University of São Paulo, Brazil · HUG Labs, Brazil · Institute of Informatics, Federal University of Goiás, Brazil +1
Sampling for Quality: Training-Free Reward-Guided LLM Decoding via Sequential Monte Carlo
Apr 7, 2026Jelena Markovic-Voronov, Wenhui Zhu, Bo Long +5Language Model Decoding LLM Code Generation
LinkedIn
Inference-Time Code Selection via Symbolic Equivalence Partitioning
Apr 7, 2026David Cho, Yifan Wang, Fanping Sui +1LLM Code Generation Inference-Time Search
Department of Computer Science, Purdue University, West Lafayette, IN, USA · Texas Instruments, Dallas, TX, USA
LLM4CodeRE: Generative AI for Code Decompilation Analysis and Reverse Engineering
Apr 7, 2026Hamed Jelodar, Samita Bai, Tochukwu Emmanuel Nwankwo +4Software Reverse Engineering Malware Analysis
Faculty of Computer Science, University of New Brunswick, Canada
QiMeng-PRepair: Precise Code Repair via Edit-Aware Reward Optimization
Apr 7, 2026Changxin Ke, Rui Zhang, Jiaming Guo +10Automated Program Repair LLM Code Generation
State Key Lab of Processors, Institute of Computing Technology, CAS · University of Chinese Academy of Sciences · Institute of Microelectronics, CAS
Broken by Default: A Formal Verification Study of Security Vulnerabilities in AI-Generated Code
Apr 7, 2026Dominik Blain, Maxime NoiseuxLLM Security LLM Code Generation
Cobalt AI Gatineau, QC, Canada
Compiled AI: Deterministic Code Generation for LLM-Based Workflow Automation
Apr 6, 2026Geert Trooskens, Aaron Karlsberg, Anmol Sharma +6Code Generation LLM Code Generation
XY.AI Labs, Palo Alto, CA · Stanford University School of Medicine, Stanford, CA · Cornell University, Ithaca, NY +3
EffiPair: Improving the Efficiency of LLM-generated Code with Differential Execution Feedback
Apr 6, 2026Samira Hajizadeh, Suman JanaInference-Time Optimization Code Optimization
Department of Electrical Engineering, Columbia University, New York, NY, USA · Department of Computer Science, Columbia University, New York, NY, USA
From Paper to Program: Specification-Guided Implementation of Quantum Many-Body Algorithms with Large Language Model
Apr 5, 2026Yi ZhouScientific Code Generation Code Generation Evaluation
Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China
SecPI: Secure Code Generation with Reasoning Models via Security Reasoning Internalization
Apr 4, 2026Hao Wang, Niels Mündler, Mark Vero +3Secure Code Generation LLM Code Generation
University of California, Berkeley · ETH Zurich
InCoder-32B-Thinking: Industrial Code World Model for Thinking
Apr 3, 2026Jian Yang, Wei Zhang, Jiajun Wu +22LLM Code Generation Code Generation
Beihang University · IQuest Research · Shanghai Jiao Tong University +5
From Theory to Practice: Code Generation Using LLMs for CAPEC and CWE Frameworks
Apr 2, 2026Murtuza Shahzad, Joseph Wilson, Ibrahim Al Azher +2Software Security Synthetic Data Generation for Language Models
Department of Computer Science, Northern Illinois University, DeKalb, Illinois USA
From Guessing to Placeholding: A Cost-Theoretic Framework for Uncertainty-Aware Code Completion
Apr 2, 2026Liang Zhu, Haolin Chen, Lidong Zhao +1Code Generation LLM Uncertainty Estimation
1Tencent, Shenzhen, China
Automating Database-Native Function Code Synthesis with LLMs
Apr 2, 2026Wei Zhou, Xuanhe Zhou, Qikang He +4LLM-Based Program Synthesis LLM Code Generation
Shanghai Jiao Tong University, China · Tsinghua University, China · National University of Singapore, Singapore +1
Embarrassingly Simple Self-Distillation Improves Code Generation
Apr 1, 2026Ruixiang Zhang, Richard He Bai, Huangjie Zheng +3Language Model Decoding Language Model Distillation
Apple

Apr 17, 2026J/K move · Enter open · S save

ACE: Self-Evolving LLM Coding Framework via Adversarial Unit Test Generation and Preference Optimization

Yixu Huang, Xinglei Yu, Zhongyu Wei

School of Data Science Fudan University Shanghai, China

Execution-Guided Code Generation Language Model Self-Improvement Automated Test Generation LLM Code Generation

Open paper AskPDF arXiv

Large Language Models (LLMs) excel at code generation but remain heavily reliant on large-scale annotated solutions and verification-based supervision, which constrains scalability and hinders sustained self-improvement. Recent solver--verifier frameworks exploit program execution as an automatic supervision signal, but their effectiveness degrades as solvers become moderately strong: verifier-generated tests increasingly confirm semantic correctness rather than exposing the remaining failure modes. We propose ACE, a self-evolving code generation framework based on a solver--adversary architecture that prioritizes active failure discovery through execution-centric supervision. A single LLM alternates between generating candidate programs and producing adversarial unit test inputs optimized to induce execution-level failures, such as runtime errors, exceptions, or non-termination. Supervision is derived solely from execution outcomes: robust programs are selected for supervised fine-tuning, while adversarial tests are optimized via Kahneman--Tversky Optimization using execution-derived preferences. Notably, the entire training loop requires no ground-truth code or external reward models. Experiments on CodeContests, MBPP, and LiveCodeBench demonstrate that ACE consistently outperforms strong solver--verifier baselines, achieving 3--7% absolute gains in pass@1, with larger improvements on out-of-distribution benchmarks, while maintaining competitive or improved inference efficiency.