中文
相关论文

相关论文: Guardrails Beat Guidance: A Large-Scale Study of R…

200 篇论文

Agent skills extend large language model (LLM) agents with reusable, program-like modules that define triggering conditions, procedural logic, and tool interactions. As these skills proliferate in public marketplaces, it is unclear what…

软件工程 · 计算机科学 2026-02-10 George Ling , Shanshan Zhong , Richard Huang

Coding agents represent a new paradigm in automated software engineering, combining the reasoning capabilities of Large Language Models (LLMs) with tool-augmented interaction loops. However, coding agents still have severe limitations.…

软件工程 · 计算机科学 2026-04-06 Tural Mehtiyev , Wesley Assunção

Agentic systems are evaluated on benchmarks where agents interact with environments to solve tasks. Most papers report a pass@1 score computed from a single run per task, assuming this gives a reliable performance estimate. We test this…

机器学习 · 计算机科学 2026-03-26 Bjarni Haukur Bjarnason , André Silva , Martin Monperrus

Translating statistical methods into reliable software is a persistent bottleneck in quantitative research. Existing AI code-generation tools produce code quickly but cannot guarantee faithful implementation -- a critical requirement for…

软件工程 · 计算机科学 2026-04-07 Tianzhu Qin , Yiqing Xu

Recent advancements in Large Language Models (LLMs) have spurred interest in deploying LLM agents to undertake tasks in the world. LLMs are often deployed in agent systems: code that orchestrates LLM calls and provides them with tools. We…

人工智能 · 计算机科学 2025-05-20 Maxime Robeyns , Martin Szummer , Laurence Aitchison

Recent developments in sequential experimental design look to construct a policy that can efficiently navigate the design space, in a way that maximises the expected information gain. Whilst there is work on achieving tractable policies for…

机器学习 · 计算机科学 2025-08-20 Yasir Zubayr Barlas , Kizito Salako

AI coding agents powered by large language models can read codebases and produce functional code, but they routinely violate team-specific product decisions that are invisible in the source code alone. We introduce a controlled benchmark…

软件工程 · 计算机科学 2026-05-12 Drew Dillon , Kasyap Varanasi

Computer-use agents (CUAs) can now autonomously complete complex tasks in real digital environments, but when misled, they can also be used to automate harmful actions programmatically. Existing safety evaluations largely target explicit…

密码学与安全 · 计算机科学 2026-04-20 Xuwei Ding , Skylar Zhai , Linxin Song , Jiate Li , Taiwei Shi , Nicholas Meade , Siva Reddy , Jian Kang , Jieyu Zhao

Enterprise teams building internal coding agents face a gap between prototype performance and production readiness. The root cause is that technical model quality alone is insufficient -- tool design, safety enforcement, state management,…

软件工程 · 计算机科学 2026-04-14 Gustavo Pinto , Pedro Eduardo de Paula Naves , Ana Paula Camargo , Marselle Silva

Agentic coding tools, such as OpenAI Codex, Claude Code, and Cursor, are transforming the software engineering landscape. These AI-powered systems function as autonomous teammates capable of planning and executing complex development tasks.…

软件工程 · 计算机科学 2025-11-10 Kosei Horikawa , Hao Li , Yutaro Kashiwa , Bram Adams , Hajimu Iida , Ahmed E. Hassan

Coding agents produce rich trajectories while solving software-engineering tasks. To enable agent self-evolution, these trajectories can be distilled into reusable procedural skills that compactly encode experience to guide future behavior.…

人工智能 · 计算机科学 2026-05-26 Yanzhou Li , Yiran Zhang , Xiaoyu Zhang , Xiaoxia Liu , Yang Liu

Street-level bureaucrats, such as caseworkers and border guards routinely face the dilemma of whether to follow rigid policy or exercise discretion based on professional judgement. However, frequent overrides threaten consistency and…

计算机与社会 · 计算机科学 2026-02-11 Gaurab Pokharel , Sanmay Das , Patrick J. Fowler

Automated program repair has traditionally focused on single-hunk defects, overlooking multi-hunk bugs that are prevalent in real-world systems. Repairing these bugs requires coordinated edits across multiple, disjoint code regions, posing…

软件工程 · 计算机科学 2025-11-17 Noor Nashid , Daniel Ding , Keheliya Gallaba , Ahmed E. Hassan , Ali Mesbah

Agent Skills, structured packages of procedural knowledge loaded into an LLM agent at inference time, are widely reported to improve task pass rates by an average of 16.2~percentage points across diverse domains. Yet the same benchmarks…

人工智能 · 计算机科学 2026-05-26 Samuel Jacob Chacko , James Hugglestone , Chashi Mahiul Islam , Xiuwen Liu

During training, reinforcement learning systems interact with the world without considering the safety of their actions. When deployed into the real world, such systems can be dangerous and cause harm to their surroundings. Often, dangerous…

人工智能 · 计算机科学 2022-12-29 Ekaterina Nikonova , Cheng Xue , Jochen Renz

Scaling up high-quality trajectory data has long been a critical bottleneck for developing human-like computer use agents. We introduce PC Agent-E, an efficient agent training framework that significantly reduces reliance on large-scale…

人工智能 · 计算机科学 2026-03-04 Yanheng He , Jiahe Jin , Pengfei Liu

Claude Code's auto mode is the first deployed permission system for AI coding agents, using a two-stage transcript classifier to gate dangerous tool calls. Anthropic reports a 0.4% false positive rate and 17% false negative rate on…

软件工程 · 计算机科学 2026-04-29 Zimo Ji , Zongjie Li , Wenyuan Jiang , Yudong Gao , Shuai Wang

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…

软件工程 · 计算机科学 2026-03-30 Yangtian Zi , Zixuan Wu , Aleksander Boruch-Gruszecki , Jonathan Bell , Arjun Guha

As agent systems scale, skills accumulate into large reusable libraries, yet their scaling laws remain poorly understood. Across 15 frontier LLMs, 1,141 real-world skills, and over 3M routing or execution decisions, we identify two coupled…

Large language model (LLM) based coding agents increasingly act as autonomous contributors that generate and merge pull requests, yet their real-world effects on software projects are unclear-especially compared with widely adopted…

软件工程 · 计算机科学 2026-01-28 Shyam Agarwal , Hao He , Bogdan Vasilescu