CAS and Tsinghua University Collaborate to Create a Programming Darwin: SE-Agent Self-Evolution Intelligent Agent Surpasses 61.2% Success Rate, Open Source Sparks Global Excitement
A quiet programming revolution is unfolding in the world of code. While traditional AI is still solving problems through brute force, an autonomous evolutionary intelligent agent called SE-Agent has learned to think like biological evolution. Every line of code is self-optimized, and each attempt accumulates wisdom. This innovative framework, jointly developed by top institutions such as CAS, Tsinghua University, and StepZen, is redefining artificial intelligence's programming capabilities in an unprecedented way. SE-Agent has successfully broken through the programming bottlenecks of the Claude-4 model,