Mem-$\pi$: Adaptive Memory through Learning When and What to Generate

ArXi:2605.21463v1 Announce Type: new We present Mem-$\pi$, a framework for adaptive memory in large language model (LLM) agents, where useful guidance is generated on demand rather than retrieved from external memory s. Existing memory-augmented agents typically rely on similarity-based retrieval from episodic memory banks or skill libraries, returning static entries that often misalign with the current context.