中文
相关论文

相关论文: SWE-TRACE: Optimizing Long-Horizon SWE Agents Thro…

200 篇论文

Automating real-world software engineering tasks remains challenging for large language model (LLM)-based agents due to the need for long-horizon reasoning over large, evolving codebases and making consistent decisions across interdependent…

软件工程 · 计算机科学 2026-04-14 Mahir Labib Dihan , Md Ashrafur Rahman Khan

Execution-based feedback like unit testing is widely used in the development of coding agents through test-time scaling (TTS) and reinforcement learning (RL). This paradigm requires scalable and reliable collection of unit test cases to…

计算与语言 · 计算机科学 2025-12-29 KaShun Shum , Binyuan Hui , Jiawei Chen , Lei Zhang , X. W. , Jiaxi Yang , Yuzhen Huang , Junyang Lin , Junxian He

LLM-based agents have shown promising capabilities in a growing range of software engineering (SWE) tasks. However, advancing this field faces two critical challenges. First, high-quality training data is scarce, especially data that…

Software engineers operating in complex and dynamic environments must continuously adapt to evolving requirements, learn iteratively from experience, and reconsider their approaches based on new insights. However, current large language…

人工智能 · 计算机科学 2025-04-03 Antonis Antoniades , Albert Örwall , Kexun Zhang , Yuxi Xie , Anirudh Goyal , William Wang

Test-time scaling has been widely adopted to enhance the capabilities of Large Language Model (LLM) agents in software engineering (SWE) tasks. However, the standard approach of repeatedly sampling trajectories from scratch is…

软件工程 · 计算机科学 2026-02-06 Yifeng Ding , Lingming Zhang

Recent advancements in software engineering agents have demonstrated promising capabilities in automating program improvements. However, their reliance on closed-source or resource-intensive models introduces significant deployment…

软件工程 · 计算机科学 2025-04-09 Yingwei Ma , Yongbin Li , Yihong Dong , Xue Jiang , Rongyu Cao , Jue Chen , Fei Huang , Binhua Li

In this technical report, we present SWE-Master, an open-source and fully reproducible post-training framework for building effective software engineering agents. SWE-Master systematically explores the complete agent development pipeline,…

Although recent tool-augmented benchmarks involve complex requests, evaluation remains limited to answer matching, neglecting critical trajectory aspects like efficiency, hallucination, and adaptivity. The most straightforward method for…

人工智能 · 计算机科学 2026-05-26 Wonjoong Kim , Sangwu Park , Yeonjun In , Sein Kim , Dongha Lee , Chanyoung Park

Large Language Model (LLM)-based agents have recently shown impressive capabilities in complex reasoning and tool use via multi-step interactions with their environments. While these agents have the potential to tackle complicated tasks,…

Large Language Model (LLM) agents are increasingly deployed for complex, multi-step software engineering (SWE) tasks. However, their trajectories often contain costly inefficiencies, such as redundant exploration, looping, and failure to…

人工智能 · 计算机科学 2025-10-22 Shubham Gandhi , Jason Tsay , Jatin Ganhotra , Kiran Kate , Yara Rizk

LLM agents have demonstrated remarkable capabilities in software development, but their performance is hampered by long interaction contexts, which incur high API costs and latency. While various context compression approaches such as…

软件工程 · 计算机科学 2026-05-08 Yuhang Wang , Yuling Shi , Mo Yang , Rongrui Zhang , Shilin He , Heng Lian , Yuting Chen , Siyu Ye , Kai Cai , Xiaodong Gu

As autonomous code agents move toward end-to-end software development, evaluating their practical autonomy becomes critical. Current benchmarks hide friction by testing agents in pre-configured environments, and their static evaluation…

软件工程 · 计算机科学 2026-05-14 Hao Guan , Lingyue Fu , Shao Zhang , Yaoming Zhu , Kangning Zhang , Lin Qiu , Xunliang Cai , Xuezhi Cao , Weiwen Liu , Weinan Zhang , Yong Yu

Large language models (LLMs) exhibit strong performance on self-contained programming tasks. However, they still struggle with repository-level software engineering (SWE), which demands (1) deep codebase navigation with effective context…

软件工程 · 计算机科学 2026-05-27 Kang He , Kaushik Roy

Despite recent progress in Large Language Model (LLM) Agents for Software Engineering (SWE) tasks, end-to-end fine-tuning typically relies on verifiable terminal rewards such as whether all unit tests pass. While these binary signals…

机器学习 · 计算机科学 2026-04-21 Jiawei Huang , Qingping Yang , Renjie Zheng , Jiaze Chen

The evaluation of Deep Research Agents is a critical challenge, as conventional outcome-based metrics fail to capture the nuances of their complex reasoning. Current evaluation faces two primary challenges: 1) a reliance on singular metrics…

计算与语言 · 计算机科学 2026-02-26 Yanyu Chen , Jiyue Jiang , Jiahong Liu , Yifei Zhang , Xiao Guo , Irwin King

Recent advances in large language models (LLMs) and agent system designs have empowered agents with unprecedented levels of capability. However, existing agent benchmarks are showing a trend of rapid ceiling-hitting by newly developed…

人工智能 · 计算机科学 2026-03-25 Dadi Guo , Tianyi Zhou , Dongrui Liu , Chen Qian , Qihan Ren , Shuai Shao , Zhiyuan Fan , Yi R. Fung , Kun Wang , Linfeng Zhang , Jing Shao

Software engineering agents (SWE) are improving rapidly, with recent gains largely driven by reinforcement learning (RL). However, RL training is constrained by the scarcity of large-scale task collections with reproducible execution…

软件工程 · 计算机科学 2026-03-02 Ibragim Badertdinov , Maksim Nekrashevich , Anton Shevtsov , Alexander Golubev

Prior representative ReAct-style approaches in autonomous Software Engineering (SWE) typically lack the explicit System-2 reasoning required for deep analysis and handling complex edge cases. While recent reasoning models demonstrate the…

人工智能 · 计算机科学 2026-04-14 Shuquan Lian , Juncheng Liu , Yazhe Chen , Yuhong Chen , Hui Li

Auto-regressive LLM-based software engineering (SWE) agents, henceforth SWE agents, have made tremendous progress (>60% on SWE-Bench Verified) on real-world coding challenges including GitHub issue resolution. SWE agents use a combination…

软件工程 · 计算机科学 2025-04-15 Timothy Bula , Saurabh Pujar , Luca Buratti , Mihaela Bornea , Avirup Sil

Optimizing the performance of large-scale software repositories demands expertise in code reasoning and software engineering (SWE) to reduce runtime while preserving program correctness. However, most benchmarks emphasize what to fix rather…

‹ 上一页 1 2 3 10 下一页 ›