Researchers have developed EvoOtter, a novel approach that combines evolutionary programming with large language models to generate bug reproduction tests (BRTs) more efficiently. This method aims to reduce the costs associated with generating and executing tests, which are often high due to the informal nature of issue descriptions. EvoOtter controls execution costs through successive halving and LLM costs via batched crossover and rule-based mutations, utilizing a new fitness score tailored for BRTs. AI
IMPACT This approach could significantly reduce the cost and improve the effectiveness of debugging software by leveraging LLMs and evolutionary programming.
RANK_REASON The item describes a new research paper detailing a novel method for software engineering. [lever_c_demoted from research: ic=1 ai=0.7]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
