Latest AI for Code Research Papers

The newest AI for Code papers from across the field — arXiv, NeurIPS, CVPR, Nature, and more — refreshed daily and ranked by relevance. Distill AI tracks AI for Code so you don’t have to: get the standout work delivered to your inbox every morning, with 2-sentence summaries and the option to chat with any paper.

Recent papers

• REALMS-CG: Resilient Ensemble of Autonomous LLM Multi-agent with Spectral clustering for Code Generation
  Anonymous, Anonymous · Zenodo (CERN European Organ... · Sep 1, 2027

  This contains the actual source code, prompts, scripts and all experiment results as described in the paper:REALMS-CG: Resilient Ensemble of Autonomous LLM Multi-agent with Spectral clustering for Code Generation.This data is provided as pa…

• When Requirements Interact: Scaling Long-Horizon Coding via Interdependent Specifications
  Yishuo Cai, Minzheng Wang, Liang Zhao, Linghao Zhang et al. · arXiv · Oct 29, 2026

  To succeed in real-world long-horizon software development, coding agents must satisfy multi-dimensional task specifications, such as interface compatibility and error handling, and do so without breaking existing dependencies. However, cur…

• PR4Code: A Pull Requests Dataset for AI Code Generation
  Benedetta Donato, Leonardo Mariani, Daniela Micucci, Oliviero Riganelli · arXiv · Oct 29, 2026

  AI-driven code generation is evolving from handling small, localized code-completion tasks toward supporting the implementation of full software features. To support this transition, this paper introduces PR4Code, a dataset of 4,508 Java an…

• Soei: Structured Observability of Evidence and Intent for Multi-Agent AI Software Engineering
  Mohammed Nayeem Khan, K S Sarun Kumar · arXiv · Oct 29, 2026

  AI coding agents now do long, multi-step and multi-session software work with little supervision. When a developer comes back to that work, the useful question is not what the model said but what was actually done, by whom, and whether it i…

• Well-Specified and Well-Tested? Assessing SWE-bench-Style Datasets with a Self-Evolving Agent
  Zhonghao Jiang, Xuyang Zheng, Yanfu Yan, Zhongxin Liu et al. · arXiv · Oct 29, 2026

  Large language models (LLMs) have shown strong performance on software engineering (SE) tasks, driving the need for high-quality datasets for reliable training and evaluation. SWE-bench-style datasets (SWE datasets) are widely used to evalu…

• When Relevant Code Is Wrong: Runtime-Aware Context for AI Coding Agents
  Wesley K. G. Assunção, Ramesh Periyathambi · arXiv · Oct 28, 2026

  AI coding agents increasingly rely on retrieved code and development history to guide software changes. However, highly relevant artifacts may describe implementations that are inactive or incompatible with the target deployment. Neither se…

• CARI: Cross-Boundary Adversarial Runtime Injection in Coding-Agent Workflows
  Jie Fu, Shuang Wu, Ashish Kundu, Wendy Hui Wang et al. · arXiv · Oct 28, 2026

  Coding agents increasingly develop, test, and integrate software whose components are owned by different developers, teams, or services. This exposes a distinct attack surface: a malicious or compromised peer development role can manipulate…

• How Deterministic SAST and AI Security Agents Divide the Work: Coverage, Harness Policy and Cost in Snyk VulnBench 2.0
  Liran Tal, Chris Liu, Charlotte Fereday, Sebastian Roth et al. · arXiv · Oct 28, 2026

  We ask what AI coding agents and AI security harnesses find that deterministic static application security testing (SAST) does not, and what they miss that SAST catches. VulnBench 2.0 is a benchmark of 108 hand-curated, attacker-reachable v…

• Security Impact of Coding Agent's Memory Files
  Thiago Santos de Moura, Selma Güler, Yannic Noller · arXiv · Oct 28, 2026

  Coding agents automatically load project memory files such as CLAUDE.md at the start of every session and treat them as trusted project guidance. Anyone who can change such a file, for example through a pull request, can therefore steer the…

• FeatureArc: Benchmarking Coding Agents on Evolving Features
  Yunlong Wang, Alyssa Wu, Cuiyun Gao, Jia Feng et al. · arXiv · Oct 27, 2026

  A software feature usually evolves over time: developers extend its first version as new use cases and requirements arrive, and each extension builds on code written earlier. Most feature benchmarks evaluate a single request on a fixed code…

• Hidden in Plain Sight: an Empirical Study about The Instruction Stacks of Coding Agents
  Felipe de Sant'Anna Paixão, Dhyego Tavares, Eduardo Almeida, Iftekhar Ahmed · arXiv · Oct 27, 2026

  Coding agents send their models more than the user's prompt: they add instructions, tool definitions, and environmental context that form an instruction stack. This stack guides execution and changes across releases, yet remains largely une…

• Spine: Compiling Subtask Trees into Structured Context for Long-Horizon Coding Agents
  Jiahong Xiang, Kunqiu Chen, Lingming Zhang, Yuqun Zhang · arXiv · Oct 27, 2026

  Long-horizon coding tasks contain nested subtasks. Typically, agents record interactions sequentially into a growing history, causing context inflation and performance degradation. Compressing or trimming the history without explicit subtas…

• Does More Thinking Pay Off? Reasoning Cost and Correctness of LLMs in Code Generation and Bug Fixing
  Andie Huang, Ying Pang, Qian Zhang, Yihang Cheng et al. · arXiv · Oct 27, 2026

  Many LLMs now emit an explicit reasoning trace before they write code, but benchmark results for code generation and bug fixing usually report correctness alone. It is therefore unclear how much the extra reasoning contributes relative to t…

• A lightweight C Code Generator for Heterogeneous TinyML Systems
  Chen Liu, Bernhard Vogginger, Christian Mayr · arXiv · Oct 27, 2026

  Over the past decade, the significant advancements in artificial intelligence have been closely tied to increasingly sophisticated training frameworks and steadily powerful computing hardware. However, the entire ecosystem remains fragmente…

• Benchmarking LLM Agents for Behaviorally Adequate Test Generation from Software Requirements
  Binhang Qi, Yiming Liu, Yun Lin, Yangjie Zhou et al. · arXiv · Oct 27, 2026

  As large language model agents increasingly compile natural-language requirements into repository-level code, requirement-to-test generation grows in importance to provide validation constraints (or harness) to guide code generation. Yet it…

• From Ripple to Boundary: Deterministic Impact Boundaries for LLM-Based Software Evolution
  Jiameng Bai, Zhitao Wang, Long Wang, Keyue Qian et al. · arXiv · Oct 27, 2026

  The landscape of software evolution is undergoing a fundamental paradigm shift in the era of Large Language Models. As industrial software engineering increasingly transitions toward design-driven code generation, the bottleneck of evolutio…

• kw-bench: Breaking Frontier Harnesses on Real-World Tasks
  George Juraj Salapa · arXiv · Oct 26, 2026

  Each new generation of frontier model pushes the edge of what is possible. Inside a harness -- a loop, a shell, a filesystem, the ability to write and run its own code -- a model can take on increasingly difficult work. A growing body of wo…

• Your Wish Is My Commit: Over-Compliant Refactoring in AI Coding Agents
  Haocong Jiang, Sadra Sabouri, Sepand Haghighi, Souti Chattopadhyay · arXiv · Oct 26, 2026

  Large language models are known to be sycophantic, over-complying with users even when doing so is unproductive. When these models act as coding agents, this can be costly: an agent that keeps editing after no edit improves the code writes …

• How Stable Are LLM-Generated Proofs Under Program Refactoring? A Stress Test for Verifiable Code Generation
  Aravind Mandiga, Guoming Li · arXiv · Oct 26, 2026

  Benchmarks for verifiable code generation usually test a proof against one fixed implementation. We ask whether the same proof remains valid after a program refactoring that preserves behavior. We introduce RefactorProof, which derives 458 …

• Instant Software Factories
  Wenxin Jiang, Mikola Lysenko · arXiv · Oct 26, 2026

  Coding agents can quickly build prototypes, but a long tail of fixes, optimizations, tests, and releases separates a prototype from a product. A software factory handles this ongoing work through a repeatable process. We ask whether this wo…

• Agent or Human? Detecting Commits by AI Coding Agents in the Absence of Authorship Signatures
  Kosei Horikawa, Hao Li, Yutaro Kashiwa, Youssef Esseddiq Ouatiti et al. · arXiv · Oct 26, 2026

  AI coding agents such as Claude Code, GitHub Copilot coding agent, and Cursor now commit directly to open-source repositories. Empirical studies of these agents identify their commits through authorship signatures, such as bot accounts and …

• Stratum: Repository Memory for Coding Agents Beyond the Context Window
  Yuvraj Sehgal, Mahsa Panahandeh, Naser Ezzati-Jivan · arXiv · Oct 26, 2026

  Context windows have grown a thousand-fold, yet coding agents still start every session with no memory of their past work. They re-read files, retry failed approaches, and forget root causes they established hours earlier. This is not a con…

• AGDecode: Semantic Attribute Grammar Derivatives for Safety-Constrained Decoding of Large Language Models
  Rajan Maghera, Asher Tempest Arya, Aren A. Babikian, Marsha Chechik et al. · arXiv · Oct 26, 2026

  Large language models (LLMs) are increasingly used for code generation, motivating techniques that guarantee the structural and semantic correctness of generated programs. Constrained decoding enforces correctness during generation by restr…

• Do Coding Agents Still Need a Debugger? Frontier-Model-Based Fault Localization at the Accuracy Ceiling
  Christoph Daxerer, Johann Glock, Christian Macho, Rudolf Ramler et al. · arXiv · Oct 26, 2026

  Background: To localize a fault, a coding agent can reason from source code alone or use debugger assistance by running the program and stepping through it in an interactive debugger. While earlier agents struggled with static execution tra…

• GEAR: Feature-Based Configuration and Code Generation for Agentic Systems
  Brell Sanwouo, Nada Zine, Paul Temple, Clément Quinton et al. · arXiv · Oct 26, 2026

  Engineering multi-agent systems (MAS) requires choosing agents, orchestration strategies, communication mechanisms, reasoning approaches, and memory configurations. These choices depend on framework-specific capabilities and abstractions, r…

• Ranking Function Generation with LLMs for Termination
  Rongkang Li, Jiaxin Hu, Taolue Chen, Yong Li et al. · arXiv · Oct 26, 2026

  Ranking functions are a fundamental technique for proving loop termination, but constructing them manually is challenging and requires substantial expertise. We propose RAINFALL, a neuro-symbolic framework which follows the generate-verify-…

• FeatureLoop: Benchmarking Coding Agents under Evolving Requirements
  Jia Feng, Alyssa Wu, Cuiyun Gao, Yunlong Wang et al. · arXiv · Oct 26, 2026

  Software development with coding agents often begins with a rough request that users refine after trying the implementation. Final correctness alone does not capture the repeated guidance this process requires. We introduce FeatureLoop, a b…

• Making Small Code LLMs More Efficient for Secure Code Review with Neuro-Symbolic Fine-Tuning
  Mostafijur Rahman Akhond, Mahmuda Khatun, Gias Uddin · arXiv · Oct 26, 2026

  As AI coding assistants generate more code, the need to review it for security flaws grows. Existing approaches leave a gap: static analyzers identify program sites but offer little explanation, while language-model explanations can be diff…

• Toward Evidence-Based Agentic Software Engineering
  Sebastian Baltes, Christoph Treude · arXiv · Oct 26, 2026

  Task success rates and execution costs of agentic coding tools depend on both model and harness. The harness manages context, provides tools, and runs the agent loop. In empirical software engineering (SE) studies, comparisons that change m…

• Testing with Coding Agents: A Large-Scale Comparison of Agent-Supported and Human-Authored Pull Requests
  Roberto Milanese, Francesco Salzano, Hovhannes Muradyan, Jingjing Wu et al. · arXiv · Oct 26, 2026

  Context. Coding agents increasingly contribute tests through pull requests that developers then maintain, so how testing work is divided affects long-term maintainability. Objective. We compare pull requests in which developers and coding a…