Tag
This paper introduces MemProbe, a cognitive-science-inspired framework for evaluating stability-plasticity tradeoffs in agent memory systems through experimental paradigms, providing interpretable profiles of memory maintenance over time.
UniMem proposes a self-routing framework that combines episodic and parametric memory for LLM agents, enabling adaptive memory management in boundary-agnostic task streams without task labels.
This paper investigates the stability-plasticity dilemma in reinforcement learning under gradual non-stationarity, finding that stabilizing successor features via synaptic consolidation across multiple timescales outperforms plasticity-focused methods.