AIM: Agentic Idea Manager for automated idea-driven research (arXiv:2609.38445v1)
A new arXiv paper introduces the Agentic Idea Manager (AIM), a fully autonomous framework for organizing and exploring research directions in idea-driven automated research. AIM combines an Agentic Surrogate and Agentic Acquisition (inspired by Bayesian optimization), plus a Solution Auditor and Resource Planner; on 10 AutoLab benchmark tasks it improves over the strongest baseline by 1.6 percentage points on System Optimization tasks and 4.9 points on long-horizon Model Development & CUDA tasks, and reaches best-baseline performance up to 3.1x faster in wall-clock time.