中文
相关论文

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

200 篇论文

Software development, often perceived as a technical endeavor, is fundamentally a social activity requiring collaboration among team members. Acknowledging this, the software development community has devised strategies to address possible…

软件工程 · 计算机科学 2024-09-04 Mikhail Evtikhiev , Ekaterina Koshchenko , Vladimir Kovalenko

Large Language Model (LLM) agents are increasingly proposed to automate offensive security tasks, with recent studies reporting near human-level success rates in Capture-the-Flag (CTF) challenges. We here revisit these results, providing a…

密码学与安全 · 计算机科学 2026-05-22 Youness Bouchari , Matteo Boffa , Marco Mellia , Idilio Drago , Thanh Minh Bui , Dario Rossi

Animals exhibit an innate ability to learn regularities of the world through interaction. By performing experiments in their environment, they are able to discern the causal factors of variation and infer how they affect the world's…

机器学习 · 计算机科学 2021-08-10 Sumedh A. Sontakke , Arash Mehrjou , Laurent Itti , Bernhard Schölkopf

Context: In the workplace, some individuals engage in the voluntary and intentional generation, promotion, and realization of new ideas for the benefit of individual performance, group effectiveness, or the organization. The literature…

软件工程 · 计算机科学 2016-12-15 Cleviton Monteiro , Fabio Queda Bueno da Silva , Luiz Fernando Capretz

Preference-based reward learning is widely used for shaping agent behavior to match a user's preference, yet its sparse binary feedback makes it especially vulnerable to causal confusion. The learned reward often latches onto spurious…

人工智能 · 计算机科学 2026-03-06 Minjune Hwang , Yigit Korkmaz , Daniel Seita , Erdem Bıyık

Classical coding-theoretic guarantees often rely on trust assumptions, such as requiring sufficiently many honest nodes compared with adversarial ones. These assumptions are difficult to enforce in open decentralized systems where…

信息论 · 计算机科学 2026-05-12 Hanzaleh Akbari Nodehi , Parsa Moradi , Mohammad Ali Maddah-Ali

We study what actually works and what doesn't for training large language models as agents via multi-turn reinforcement learning. Despite rapid progress, existing frameworks and definitions are fragmented, and there is no systematic…

机器学习 · 计算机科学 2025-12-09 Ruiyi Wang , Prithviraj Ammanabrolu

Software development is iterative, yet agentic coding benchmarks hide design issues through their single-shot setup. Recent iterative benchmarks attempt to remedy this but heavily constrain an agent's design decision space, making it…

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…

人工智能 · 计算机科学 2026-05-27 James Lucassen , Adam Kaufman

Resolving team conflicts requires not only task-specific competence, but also social intelligence to find common ground and build consensus. As AI agents increasingly collaborate on complex work, they must develop coordination capabilities…

As machine learning models become more capable, they have exhibited increased potential in solving complex tasks. One of the most promising directions uses deep reinforcement learning to train autonomous agents in computer network defense…

机器学习 · 计算机科学 2023-10-23 Elizabeth Bates , Vasilios Mavroudis , Chris Hicks

As large language models (LLMs) evolve into autonomous agents that execute long-horizon workflows, invoking a high-capability model at every step becomes economically unsustainable. While model routing is effective for single-turn queries,…

Reinforcement learning (RL) algorithms, due to their reliance on external systems to learn from, require digital environments (e.g., simulators) with very simple interfaces, which in turn constrain significantly the implementation of such…

编程语言 · 计算机科学 2025-04-29 Massimo Fioravanti , Samuele Pasini , Giovanni Agosta

Quadratic programs arise in robotics, communications, smart grids, and many other applications. As these problems grow in size, finding solutions becomes more computationally demanding, and new algorithms are needed to efficiently solve…

最优化与控制 · 数学 2020-06-17 Matthew Ubl , Matthew T. Hale

Coding agents are rapidly changing the landscape of software development, moving from inline completion to autonomous systems that edit repositories, open pull requests, respond to issues, and run scheduled or webhook triggered routines…

软件工程 · 计算机科学 2026-05-11 Nghi D. Q. Bui , Georgios Evangelopoulos

In the first half of 2025, coding agents have emerged as a category of development tools that have very quickly transitioned to the practice. Unlike ''traditional'' code completion LLMs such as Copilot, agents like Cursor, Claude Code, or…

软件工程 · 计算机科学 2026-04-09 Romain Robbes , Théo Matricon , Thomas Degueule , Andre Hora , Stefano Zacchiroli

Reinforcement Learning is a highly active research field with promising advancements. In the field of autonomous driving, however, often very simple scenarios are being examined. Common approaches use non-interpretable control commands as…

机器学习 · 计算机科学 2025-05-06 Daniel Bogdoll , Jing Qin , Moritz Nekolla , Ahmed Abouelazm , Tim Joseph , J. Marius Zöllner

Insecure default values in software settings can be exploited by attackers to compromise the system that runs the software. As a countermeasure, there exist security-configuration guides specifying in detail which values are secure.…

软件工程 · 计算机科学 2023-08-03 Patrick Stöckle , Michael Sammereier , Bernd Grobauer , Alexander Pretschner

Large Language Model (LLM) Agents are advancing quickly, with the increasing leveraging of LLM Agents to assist in development tasks such as code generation. While LLM Agents accelerate code generation, studies indicate they may introduce…

软件工程 · 计算机科学 2026-01-30 Haoming Huang , Pongchai Jaisri , Shota Shimizu , Lingfeng Chen , Sota Nakashima , Gema Rodríguez-Pérez

This article presents GuideR, a user-guided rule induction algorithm, which overcomes the largest limitation of the existing methods-the lack of the possibility to introduce user's preferences or domain knowledge to the rule learning…

机器学习 · 计算机科学 2019-04-23 Marek Sikora , Łukasz Wróbel , Adam Gudyś