Researchers from Stanford University, SambaNova Systems, and UC Berkeley have proposed Agentic Context Engineering (ACE), a new framework designed to improve large language models (LLMs) through evolv

InfoQ is a leading online platform for software developers, architects, and technical leaders, providing news, articles, presentations, and interviews on a wide range of topics, including agile practices, DevOps, microservices, and emerging technologies. With a focus on quality content and expert insights, InfoQ helps professionals stay informed about the latest trends, best practices, and industry developments. Developers can learn from real-world experiences, gain  knowledge, and connect with peers in the global software community through InfoQ's diverse and engaging content.

InfoQ

Stanford University, SambaNova Systems, and UC Berkeley researchers developed ACE (Agentic Context Engineering), a framework that improves LLM performance through evolving structured contexts instead of model retraining. ACE uses three components—Generator, Reflector, and Curator—to perform incremental context updates that preserve knowledge while adding new insights. In benchmarks, ACE achieved 59.5% accuracy on AppWorld and 8.6% improvement on financial reasoning tasks, while reducing adaptation latency by 86.9% and computational costs by 75% compared to existing methods. The approach enables continuous learning through context evolution without fine-tuning, offering benefits for domains requiring transparency like finance and healthcare.

Researchers Introduce ACE, a Framework for Self-Improving LLM Contexts