ARISE toolset enhances AI agents for code fault localization and repair

Researchers have developed ARISE, a new system designed to improve the accuracy of AI agents in localizing and repairing software faults. ARISE enhances large language models by providing a detailed program graph that includes data-flow information, enabling agents to trace variable usage more precisely. Evaluations on SWE-bench Lite demonstrated significant improvements in fault localization and repair success rates compared to existing baselines. AI

IMPACT Enhances AI agent capabilities for software development, potentially improving code quality and developer productivity.

RANK_REASON This is a research paper detailing a new system and its evaluation on a benchmark. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• ARISE
• GitHub
• Python
• Qwen2.5-Coder-32B-Instruct
• SWE-bench Lite

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →