English
Related papers

Related papers: Overeager Coding Agents: Measuring Out-of-Scope Ac…

200 papers

When a phone-use agent avoids harm, does that show safety, or simply inability to act? Existing evaluations often cannot tell. A harmful outcome may be avoided because the agent recognized the risk and chose the safe action, or because it…

Fine-tuning large language models for code editing has typically relied on mining commits and pull requests. The working hypothesis has been that commit messages describe human intent in natural language, and patches to code describe the…

Software Engineering · Computer Science 2026-03-30 Yangtian Zi , Zixuan Wu , Aleksander Boruch-Gruszecki , Jonathan Bell , Arjun Guha

A frontier language model's acknowledged "helpful programming assistant" persona does not survive long agentic-coding sessions in the deployment regime that production products actually run. After hours of tool-using debugging, a model that…

Computation and Language · Computer Science 2026-05-26 Xianzhong Ding , Yangyang Yu , Changwei Liu , Bill Zhao

AI deployment in sensitive domains such as health care, credit, employment, and criminal justice is often treated as unsafe to authorize until model internals can be explained. This often leads to an excessive reliance on mechanistic…

Behavioral studies of LLM-based software engineering agents extract operational rules about which trajectory shapes correlate with higher resolution rates: that a test step follows a code modification, that error cascades are short, or that…

Software Engineering · Computer Science 2026-05-19 Wei Ma , Zhi Chen , Jingxu Gu , Tianling Li , Shangqing Liu , Lingxiao Jiang

Language Model (LM) agents for cybersecurity that are capable of autonomously identifying vulnerabilities and executing exploits have potential to cause real-world impact. Policymakers, model providers, and researchers in the AI and…

Recent advances in coding agents have shown remarkable progress in software issue resolution. In practice, real-world issues are typically bug fixes or feature requests in which human developers naturally incorporate refactoring as part of…

Software Engineering · Computer Science 2026-05-22 Zhao Tian , Zifan Zhang , Tao Xiao , Dong Wang , Masanari Kondo , Junjie Chen , Yasutaka Kamei

Workspace learning requires AI agents to identify, reason over, exploit, and update explicit and implicit dependencies among heterogeneous files in a worker's workspace, enabling them to complete both routine and advanced tasks effectively.…

Driven by Large Language Models, the single-agent, multi-tool architecture has become a popular paradigm for autonomous agents. However, this architecture introduces a severe privacy risk, which we term Tools Orchestration Privacy Risk…

Cryptography and Security · Computer Science 2026-03-09 Yuxuan Qiao , Dongqin Liu , Hongchang Yang , Wei Zhou , Songlin Hu

As AI agents surpass human capabilities, scalable oversight -- the problem of effectively supplying human feedback to potentially superhuman AI models -- becomes increasingly critical to ensure alignment. While numerous scalable oversight…

Artificial Intelligence · Computer Science 2025-04-08 Abhimanyu Pallavi Sudhir , Jackson Kaunismaa , Arjun Panickssery

AI coding scaffolds like Claude Code and Codex use retrying: blocking actions flagged as risky and continuing the trajectory. We study retrying from an AI control perspective, which treats the model as potentially adversarial. We find that…

Artificial Intelligence · Computer Science 2026-05-27 James Lucassen , Adam Kaufman

Evaluating and improving the security capabilities of code agents requires high-quality, executable vulnerability tasks. However, existing works rely on costly, unscalable manual reproduction and suffer from outdated data distributions. To…

Cryptography and Security · Computer Science 2026-05-19 Xianzhen Luo , Jingyuan Zhang , Shiqi Zhou , Rain Huang , Chuan Xiao , Qingfu Zhu , Zhiyuan Ma , Xing Yue , Yang Yue , Wencong Zeng , Wanxiang Che

Recent advances in frontier large language models have enabled code review agents that operate in open-ended, reasoning-intensive settings. However, the lack of standardized benchmarks and granular evaluation protocols makes it difficult to…

Software Engineering · Computer Science 2026-03-13 Kristen Pereira , Neelabh Sinha , Rajat Ghosh , Debojyoti Dutta

Claude Code is an agentic coding tool that can run shell commands, edit files, and call external services on behalf of the user. This study describes its comprehensive architecture by analyzing the publicly available TypeScript source code…

Software Engineering · Computer Science 2026-04-17 Jiacheng Liu , Xiaohan Zhao , Xinyi Shang , Zhiqiang Shen

Autonomous agents have rapidly matured as task executors and seen widespread deployment via harnesses such as OpenClaw. Safety concerns have rightly drawn growing research attention, and beneath them lie the values silently steering agent…

Artificial Intelligence · Computer Science 2026-05-12 Haonan Dong , Qiguan Feng , Kehan Jiang , Haoran Ye , Xin Zhang , Guojie Song

Coding agents powered by large language models are increasingly expected to perform realistic software maintenance tasks beyond isolated issue resolution. Existing benchmarks have shifted toward realistic software evolution, but they rarely…

Software Engineering · Computer Science 2026-05-15 Man Ho Lam , Chaozheng Wang , Hange Liu , Jingyu Xiao , Haau-sing Li , Jen-tse Huang , Terry Yue Zhuo , Michael R. Lyu

We stress-tested 16 leading models from multiple developers in hypothetical corporate environments to identify potentially risky agentic behaviors before they cause real harm. In the scenarios, we allowed models to autonomously send emails…

Cryptography and Security · Computer Science 2025-10-17 Aengus Lynch , Benjamin Wright , Caleb Larson , Stuart J. Ritchie , Soren Mindermann , Evan Hubinger , Ethan Perez , Kevin Troy

Frontier models are increasingly trained and deployed as autonomous agent. One safety concern is that AI agents might covertly pursue misaligned goals, hiding their true capabilities and objectives - also known as scheming. We study whether…

Artificial Intelligence · Computer Science 2025-01-16 Alexander Meinke , Bronson Schoen , Jérémy Scheurer , Mikita Balesni , Rusheb Shah , Marius Hobbhahn

The promise of AI-driven scientific discovery hinges on whether AI agents can autonomously design and execute the computational workflows that underpin modern science. Molecular dynamics (MD) simulation presents a natural test bed to…

Artificial Intelligence · Computer Science 2026-05-12 Vinay Kumar , Satyendra Rajput , Mausam , N. M. Anoop Krishnan

Vibe coding produces correct, executable code at speed, but leaves no record of the structural commitments, dependencies, or evidence behind it. Reviewers cannot determine what invariants were assumed, what changed, or why a regression…

Software Engineering · Computer Science 2026-04-21 Tianfu Wang , Zhezheng Hao , Yin Wu , Wei Wu , Qiang Lin , Hande Dong , Nicholas Jing Yuan , Hui Xiong