We present CodeNav, an LLM agent that navigates and leverages previously unseen code repositories to solve user queries. In contrast to tool-use LLM agents that require ``registration'' of all relevant tools via manual descriptions within the LLM context, CodeNav automatically indexes and searches over code blocks in the target codebase, finds relevant code snippets, imports them, and uses them to iteratively generate a solution with execution feedback. To highlight the core-capabilities of CodeNav, we first showcase three case studies where we use CodeNav for solving complex user queries using three diverse codebases. Next, on three benchmarks, we quantitatively compare the effectiveness of code-use (which only has access to the target codebase) to tool-use (which has privileged access to all tool names and descriptions). Finally, we study the effect of varying kinds of tool and library descriptions on code-use performance, as well as investigate the advantage of the agent seeing source code as opposed to natural descriptions of code. All code will be made open source under a permissive license.
Beyond Function-Level Search: Repository-Aware Dual-Encoder Code Retrieval with Adversarial Verification
International Digital Economy Academy (IDEA)(国际数字经济学院);School of Electronic and Computer Engineering, Peking University(电子与计算机工程学院,北京大学);Shenzhen International Graduate School, Tsinghua University(深圳国际研究生院,清华大学)
SWE-Chain: Benchmarking Coding Agents on Chained Release-Level Package Upgrades
The Chinese University of Hong Kong(香港中文大学);Independent(独立);ELLIS;Technical University of Darmstadt(达姆施塔特技术大学);Johns Hopkins University(约翰霍普金斯大学);Monash University(墨尔本大学)
OMPar: Automatic Parallelization with AI-Driven Source-to-Source Compilation
Ben-Gurion University(本古里安大学);Code Metal;University of Alabama at Birmingham(阿拉巴马大学伯明翰分校);Intel Labs(英特尔实验室);Cisco(思科);Technion(以色列理工学院);Stanford University(斯坦福大学)
VeriPilot: An LLM-Powered Verilog Debugging Framework
State Key Lab of Processors, Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所国家专用集成电路设计实验室);University of Chinese Academy of Sciences(中国科学院大学);North China Electric Power University(华北电力大学)
VerlTool: Towards Holistic Agentic Reinforcement Learning with Tool Use
University of Waterloo(滑铁卢大学);Sea AI Lab(Sea AI 实验室);University of Toronto(多伦多大学);Shanghai University(上海大学);HKUST(香港科技大学);M-A-P;National University of Singapore(新加坡国立大学)
Beyond Correctness: Enhancing Architectural Reasoning in Code LLMs via Scalable Labeling with Agentic Judgment
Centre for Software Excellence, Huawei Canada(华为加拿大软件卓越中心);Department of Computer Science, University of Manitoba, Canada(曼尼托巴大学计算机科学系);School of Computing, Queen’s University, Canada(皇后大学计算科学学院)
Beyond Autocomplete: Designing CopilotLens Towards Transparent and Explainable AI Coding Agents
Computer Science, University of Toronto(计算机科学,多伦多大学);University of Toronto(多伦多大学);King Abdullah University of Science and Technology(国王 Abdullah 科学与技术大学)
本文提出 CopilotLens 交互式框架,通过两级界面重建 AI 编码智能体的思考过程,将代码补全从黑盒建议转变为透明可解释的交互,以帮助开发者加深理解并校准信任。
Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search
Nanjing University(南京大学);Microsoft Research Asia(微软亚洲研究院);Peking University(北京大学);Zhejiang University(浙江大学);Wuhan University(武汉大学);Hong Kong University of Science and Technology(香港科技大学)
AgentFold: Closed-Loop Agentic Search for Protein Folding Model Design
Hunan University(湖南大学);Nanjing University(南京大学);The Chinese University of Hong Kong(香港中文大学);Zhejiang University(浙江大学);Yale University(耶鲁大学);Wuhan University of Science and Technology(武汉科技大学);Southeast University(东南大学);Stanford University(斯坦福大学)
Spider 2.0: Evaluating Language Models on Real-World Enterprise Text-to-SQL Workflows
University of Hong Kong(香港大学);Salesforce Research(Salesforce 研究院);Sea AI Lab(Sea AI 实验室);Google Deepmind(谷歌DeepMind);Google Cloud AI Research(谷歌云 AI 研究院);University of Waterloo(滑铁卢大学)